End of training
Browse files- README.md +11 -11
- adapter_1/adapter_model.safetensors +1 -1
- trainer_state.json +693 -693
README.md
CHANGED
|
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -49,16 +49,16 @@ The following hyperparameters were used during training:
|
|
| 49 |
|
| 50 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 52 |
-
| 0.
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
|
| 63 |
|
| 64 |
### Framework versions
|
|
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.2304
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 49 |
|
| 50 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 52 |
+
| 0.0416 | 1.0 | 546 | 0.2269 |
|
| 53 |
+
| 0.0243 | 2.0 | 1092 | 0.2054 |
|
| 54 |
+
| 0.0262 | 3.0 | 1638 | 0.1866 |
|
| 55 |
+
| 0.009 | 4.0 | 2184 | 0.2000 |
|
| 56 |
+
| 0.0196 | 5.0 | 2730 | 0.1928 |
|
| 57 |
+
| 0.0071 | 6.0 | 3276 | 0.2099 |
|
| 58 |
+
| 0.0054 | 7.0 | 3822 | 0.2070 |
|
| 59 |
+
| 0.0066 | 8.0 | 4368 | 0.2189 |
|
| 60 |
+
| 0.0006 | 9.0 | 4914 | 0.2325 |
|
| 61 |
+
| 0.001 | 10.0 | 5460 | 0.2304 |
|
| 62 |
|
| 63 |
|
| 64 |
### Framework versions
|
adapter_1/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 62969640
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9cf2ca9e94a0f5343ee63a5316cc4fe761e947a9d6e7ab6fb279ed9b99b3f4f5
|
| 3 |
size 62969640
|
trainer_state.json
CHANGED
|
@@ -10,1617 +10,1617 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05,
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 0.0005,
|
| 15 |
-
"loss":
|
| 16 |
"step": 25
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09,
|
| 20 |
-
"grad_norm": 0.
|
| 21 |
"learning_rate": 0.001,
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 50
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.14,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 0.
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 75
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.18,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss":
|
| 37 |
"step": 100
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.23,
|
| 41 |
-
"grad_norm": 1.
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 125
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.27,
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 150
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.32,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 0.
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 175
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.37,
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
-
"learning_rate": 0.
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 200
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.41,
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 225
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.46,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 250
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.5,
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate": 0.
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 275
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.55,
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
-
"learning_rate": 0.
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 300
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.6,
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
-
"learning_rate": 0.
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 325
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.64,
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
-
"learning_rate": 0.
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 350
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.69,
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
-
"learning_rate": 0.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 375
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.73,
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate": 0.
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 400
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.78,
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 0.
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 425
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.82,
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
-
"learning_rate": 0.
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 450
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.87,
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
-
"learning_rate": 0.
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 475
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.92,
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
-
"learning_rate": 0.
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 500
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.96,
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
-
"learning_rate": 0.
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 525
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 1.0,
|
| 160 |
-
"eval_loss": 0.
|
| 161 |
-
"eval_runtime":
|
| 162 |
-
"eval_samples_per_second": 1.
|
| 163 |
-
"eval_steps_per_second": 0.
|
| 164 |
"step": 546
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 1.01,
|
| 168 |
-
"grad_norm": 0.
|
| 169 |
-
"learning_rate": 0.
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 550
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 1.05,
|
| 175 |
-
"grad_norm": 0.
|
| 176 |
-
"learning_rate": 0.
|
| 177 |
-
"loss": 0.
|
| 178 |
"step": 575
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 1.1,
|
| 182 |
-
"grad_norm": 0.
|
| 183 |
-
"learning_rate": 0.
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 600
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 1.14,
|
| 189 |
-
"grad_norm": 0.
|
| 190 |
-
"learning_rate": 0.
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 625
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 1.19,
|
| 196 |
-
"grad_norm": 0.
|
| 197 |
-
"learning_rate": 0.
|
| 198 |
-
"loss": 0.
|
| 199 |
"step": 650
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 1.24,
|
| 203 |
-
"grad_norm": 0.
|
| 204 |
-
"learning_rate": 0.
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 675
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 1.28,
|
| 210 |
-
"grad_norm": 0.
|
| 211 |
-
"learning_rate": 0.
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 700
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 1.33,
|
| 217 |
-
"grad_norm": 0.
|
| 218 |
-
"learning_rate": 0.
|
| 219 |
-
"loss": 0.
|
| 220 |
"step": 725
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 1.37,
|
| 224 |
-
"grad_norm": 0.
|
| 225 |
-
"learning_rate": 0.
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 750
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 1.42,
|
| 231 |
-
"grad_norm": 0.
|
| 232 |
-
"learning_rate": 0.
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 775
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 1.47,
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
-
"learning_rate": 0.
|
| 240 |
-
"loss": 0.
|
| 241 |
"step": 800
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 1.51,
|
| 245 |
-
"grad_norm": 0.
|
| 246 |
-
"learning_rate": 0.
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 825
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 1.56,
|
| 252 |
-
"grad_norm": 0.
|
| 253 |
-
"learning_rate": 0.
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 850
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 1.6,
|
| 259 |
-
"grad_norm": 0.
|
| 260 |
-
"learning_rate": 0.
|
| 261 |
-
"loss": 0.
|
| 262 |
"step": 875
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 1.65,
|
| 266 |
-
"grad_norm":
|
| 267 |
-
"learning_rate": 0.
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 900
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 1.69,
|
| 273 |
-
"grad_norm": 0.
|
| 274 |
-
"learning_rate": 0.
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 925
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 1.74,
|
| 280 |
-
"grad_norm": 0.
|
| 281 |
-
"learning_rate": 0.
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 950
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 1.79,
|
| 287 |
-
"grad_norm": 0.
|
| 288 |
-
"learning_rate": 0.
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 975
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 1.83,
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
-
"learning_rate": 0.
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 1000
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 1.88,
|
| 301 |
-
"grad_norm": 0.
|
| 302 |
-
"learning_rate": 0.
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 1025
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 1.92,
|
| 308 |
-
"grad_norm": 0.
|
| 309 |
-
"learning_rate": 0.
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 1050
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 1.97,
|
| 315 |
-
"grad_norm": 0.
|
| 316 |
-
"learning_rate": 0.
|
| 317 |
-
"loss": 0.
|
| 318 |
"step": 1075
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 2.0,
|
| 322 |
-
"eval_loss": 0.
|
| 323 |
-
"eval_runtime":
|
| 324 |
-
"eval_samples_per_second": 1.
|
| 325 |
-
"eval_steps_per_second": 0.
|
| 326 |
"step": 1092
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 2.01,
|
| 330 |
-
"grad_norm": 0.
|
| 331 |
-
"learning_rate": 0.
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 1100
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 2.06,
|
| 337 |
-
"grad_norm":
|
| 338 |
-
"learning_rate": 0.
|
| 339 |
-
"loss": 0.
|
| 340 |
"step": 1125
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 2.11,
|
| 344 |
-
"grad_norm": 0.
|
| 345 |
-
"learning_rate": 0.
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 1150
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 2.15,
|
| 351 |
-
"grad_norm": 0.
|
| 352 |
-
"learning_rate": 0.
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 1175
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 2.2,
|
| 358 |
-
"grad_norm": 0.
|
| 359 |
-
"learning_rate": 0.
|
| 360 |
-
"loss": 0.
|
| 361 |
"step": 1200
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 2.24,
|
| 365 |
-
"grad_norm": 0.
|
| 366 |
-
"learning_rate": 0.
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 1225
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 2.29,
|
| 372 |
-
"grad_norm": 0.
|
| 373 |
-
"learning_rate": 0.
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 1250
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 2.34,
|
| 379 |
-
"grad_norm": 0.
|
| 380 |
-
"learning_rate": 0.
|
| 381 |
-
"loss": 0.
|
| 382 |
"step": 1275
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 2.38,
|
| 386 |
-
"grad_norm": 0.
|
| 387 |
-
"learning_rate": 0.
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 1300
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 2.43,
|
| 393 |
-
"grad_norm": 0.
|
| 394 |
-
"learning_rate": 0.
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 1325
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 2.47,
|
| 400 |
-
"grad_norm": 0.
|
| 401 |
-
"learning_rate": 0.
|
| 402 |
-
"loss": 0.
|
| 403 |
"step": 1350
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 2.52,
|
| 407 |
-
"grad_norm": 0.
|
| 408 |
-
"learning_rate": 0.
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 1375
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 2.56,
|
| 414 |
-
"grad_norm": 0.
|
| 415 |
-
"learning_rate": 0.
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 1400
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 2.61,
|
| 421 |
-
"grad_norm": 0.
|
| 422 |
-
"learning_rate": 0.
|
| 423 |
-
"loss": 0.
|
| 424 |
"step": 1425
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 2.66,
|
| 428 |
-
"grad_norm": 0.
|
| 429 |
-
"learning_rate": 0.
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 1450
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 2.7,
|
| 435 |
-
"grad_norm": 0.
|
| 436 |
-
"learning_rate": 0.
|
| 437 |
-
"loss": 0.
|
| 438 |
"step": 1475
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 2.75,
|
| 442 |
-
"grad_norm": 0.
|
| 443 |
-
"learning_rate": 0.
|
| 444 |
-
"loss": 0.
|
| 445 |
"step": 1500
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 2.79,
|
| 449 |
-
"grad_norm": 0.
|
| 450 |
-
"learning_rate": 0.
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 1525
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 2.84,
|
| 456 |
-
"grad_norm": 0.
|
| 457 |
-
"learning_rate": 0.
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 1550
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 2.88,
|
| 463 |
-
"grad_norm": 0.
|
| 464 |
-
"learning_rate": 0.
|
| 465 |
-
"loss": 0.
|
| 466 |
"step": 1575
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 2.93,
|
| 470 |
-
"grad_norm": 0.
|
| 471 |
-
"learning_rate": 0.
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 1600
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 2.98,
|
| 477 |
-
"grad_norm": 0.
|
| 478 |
-
"learning_rate": 0.
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 1625
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 3.0,
|
| 484 |
-
"eval_loss": 0.
|
| 485 |
-
"eval_runtime":
|
| 486 |
-
"eval_samples_per_second": 1.
|
| 487 |
-
"eval_steps_per_second": 0.
|
| 488 |
"step": 1638
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 3.02,
|
| 492 |
-
"grad_norm": 0.
|
| 493 |
-
"learning_rate": 0.
|
| 494 |
-
"loss": 0.
|
| 495 |
"step": 1650
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 3.07,
|
| 499 |
-
"grad_norm": 0.
|
| 500 |
-
"learning_rate": 0.
|
| 501 |
-
"loss": 0.
|
| 502 |
"step": 1675
|
| 503 |
},
|
| 504 |
{
|
| 505 |
"epoch": 3.11,
|
| 506 |
-
"grad_norm": 0.
|
| 507 |
-
"learning_rate": 0.
|
| 508 |
-
"loss": 0.
|
| 509 |
"step": 1700
|
| 510 |
},
|
| 511 |
{
|
| 512 |
"epoch": 3.16,
|
| 513 |
-
"grad_norm": 0.
|
| 514 |
-
"learning_rate": 0.
|
| 515 |
-
"loss": 0.
|
| 516 |
"step": 1725
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 3.21,
|
| 520 |
-
"grad_norm": 0.
|
| 521 |
-
"learning_rate": 0.
|
| 522 |
-
"loss": 0.
|
| 523 |
"step": 1750
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 3.25,
|
| 527 |
-
"grad_norm": 0.
|
| 528 |
-
"learning_rate": 0.
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 1775
|
| 531 |
},
|
| 532 |
{
|
| 533 |
"epoch": 3.3,
|
| 534 |
-
"grad_norm": 0.
|
| 535 |
-
"learning_rate": 0.
|
| 536 |
-
"loss": 0.
|
| 537 |
"step": 1800
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 3.34,
|
| 541 |
-
"grad_norm": 0.
|
| 542 |
-
"learning_rate": 0.
|
| 543 |
-
"loss": 0.
|
| 544 |
"step": 1825
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 3.39,
|
| 548 |
-
"grad_norm": 0.
|
| 549 |
-
"learning_rate": 0.
|
| 550 |
-
"loss": 0.
|
| 551 |
"step": 1850
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 3.43,
|
| 555 |
-
"grad_norm": 0.
|
| 556 |
-
"learning_rate": 0.
|
| 557 |
-
"loss": 0.
|
| 558 |
"step": 1875
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 3.48,
|
| 562 |
-
"grad_norm": 0.
|
| 563 |
-
"learning_rate": 0.
|
| 564 |
-
"loss": 0.
|
| 565 |
"step": 1900
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 3.53,
|
| 569 |
-
"grad_norm": 0.
|
| 570 |
-
"learning_rate": 0.
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 1925
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 3.57,
|
| 576 |
-
"grad_norm": 0.
|
| 577 |
-
"learning_rate": 0.
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 1950
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 3.62,
|
| 583 |
-
"grad_norm": 0.
|
| 584 |
-
"learning_rate": 0.
|
| 585 |
-
"loss": 0.
|
| 586 |
"step": 1975
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 3.66,
|
| 590 |
-
"grad_norm": 0.
|
| 591 |
-
"learning_rate": 0.
|
| 592 |
-
"loss": 0.
|
| 593 |
"step": 2000
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 3.71,
|
| 597 |
-
"grad_norm": 0.
|
| 598 |
-
"learning_rate": 0.
|
| 599 |
-
"loss": 0.
|
| 600 |
"step": 2025
|
| 601 |
},
|
| 602 |
{
|
| 603 |
"epoch": 3.75,
|
| 604 |
-
"grad_norm": 0.
|
| 605 |
-
"learning_rate": 0.
|
| 606 |
-
"loss": 0.
|
| 607 |
"step": 2050
|
| 608 |
},
|
| 609 |
{
|
| 610 |
"epoch": 3.8,
|
| 611 |
-
"grad_norm": 0.
|
| 612 |
-
"learning_rate": 0.
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 2075
|
| 615 |
},
|
| 616 |
{
|
| 617 |
"epoch": 3.85,
|
| 618 |
-
"grad_norm": 0.
|
| 619 |
-
"learning_rate": 0.
|
| 620 |
"loss": 0.0201,
|
| 621 |
"step": 2100
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 3.89,
|
| 625 |
-
"grad_norm": 0.
|
| 626 |
-
"learning_rate": 0.
|
| 627 |
-
"loss": 0.
|
| 628 |
"step": 2125
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 3.94,
|
| 632 |
-
"grad_norm": 0.
|
| 633 |
-
"learning_rate": 0.
|
| 634 |
-
"loss": 0.
|
| 635 |
"step": 2150
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 3.98,
|
| 639 |
-
"grad_norm": 0.
|
| 640 |
-
"learning_rate": 0.
|
| 641 |
-
"loss": 0.
|
| 642 |
"step": 2175
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 4.0,
|
| 646 |
-
"eval_loss": 0.
|
| 647 |
-
"eval_runtime":
|
| 648 |
-
"eval_samples_per_second": 1.
|
| 649 |
-
"eval_steps_per_second": 0.
|
| 650 |
"step": 2184
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 4.03,
|
| 654 |
-
"grad_norm": 0.
|
| 655 |
-
"learning_rate": 0.
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 2200
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 4.08,
|
| 661 |
-
"grad_norm": 0.
|
| 662 |
-
"learning_rate": 0.
|
| 663 |
-
"loss": 0.
|
| 664 |
"step": 2225
|
| 665 |
},
|
| 666 |
{
|
| 667 |
"epoch": 4.12,
|
| 668 |
-
"grad_norm": 0.
|
| 669 |
-
"learning_rate": 0.
|
| 670 |
-
"loss": 0.
|
| 671 |
"step": 2250
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 4.17,
|
| 675 |
-
"grad_norm":
|
| 676 |
-
"learning_rate": 0.
|
| 677 |
-
"loss": 0.
|
| 678 |
"step": 2275
|
| 679 |
},
|
| 680 |
{
|
| 681 |
"epoch": 4.21,
|
| 682 |
-
"grad_norm": 0.
|
| 683 |
-
"learning_rate": 0.
|
| 684 |
-
"loss": 0.
|
| 685 |
"step": 2300
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 4.26,
|
| 689 |
-
"grad_norm": 0.
|
| 690 |
-
"learning_rate": 0.
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 2325
|
| 693 |
},
|
| 694 |
{
|
| 695 |
"epoch": 4.3,
|
| 696 |
-
"grad_norm": 0.
|
| 697 |
-
"learning_rate": 0.
|
| 698 |
-
"loss": 0.
|
| 699 |
"step": 2350
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 4.35,
|
| 703 |
-
"grad_norm": 0.
|
| 704 |
-
"learning_rate": 0.
|
| 705 |
-
"loss": 0.
|
| 706 |
"step": 2375
|
| 707 |
},
|
| 708 |
{
|
| 709 |
"epoch": 4.4,
|
| 710 |
-
"grad_norm": 0.
|
| 711 |
-
"learning_rate": 0.
|
| 712 |
-
"loss": 0.
|
| 713 |
"step": 2400
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 4.44,
|
| 717 |
-
"grad_norm": 0.
|
| 718 |
-
"learning_rate": 0.
|
| 719 |
-
"loss": 0.
|
| 720 |
"step": 2425
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 4.49,
|
| 724 |
-
"grad_norm": 0.
|
| 725 |
-
"learning_rate": 0.
|
| 726 |
-
"loss": 0.
|
| 727 |
"step": 2450
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 4.53,
|
| 731 |
-
"grad_norm": 0.
|
| 732 |
-
"learning_rate": 0.
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 2475
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 4.58,
|
| 738 |
-
"grad_norm": 0.
|
| 739 |
-
"learning_rate": 0.
|
| 740 |
"loss": 0.0156,
|
| 741 |
"step": 2500
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 4.62,
|
| 745 |
-
"grad_norm": 0.
|
| 746 |
-
"learning_rate": 0.
|
| 747 |
-
"loss": 0.
|
| 748 |
"step": 2525
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 4.67,
|
| 752 |
-
"grad_norm": 0.
|
| 753 |
-
"learning_rate": 0.
|
| 754 |
-
"loss": 0.
|
| 755 |
"step": 2550
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 4.72,
|
| 759 |
-
"grad_norm": 0.
|
| 760 |
-
"learning_rate": 0.
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 2575
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 4.76,
|
| 766 |
-
"grad_norm": 0.
|
| 767 |
-
"learning_rate": 0.
|
| 768 |
-
"loss": 0.
|
| 769 |
"step": 2600
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 4.81,
|
| 773 |
-
"grad_norm": 0.
|
| 774 |
-
"learning_rate": 0.
|
| 775 |
-
"loss": 0.
|
| 776 |
"step": 2625
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 4.85,
|
| 780 |
-
"grad_norm": 0.
|
| 781 |
-
"learning_rate": 0.
|
| 782 |
-
"loss": 0.
|
| 783 |
"step": 2650
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 4.9,
|
| 787 |
-
"grad_norm": 0.
|
| 788 |
-
"learning_rate": 0.
|
| 789 |
-
"loss": 0.
|
| 790 |
"step": 2675
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 4.95,
|
| 794 |
-
"grad_norm": 0.
|
| 795 |
-
"learning_rate": 0.
|
| 796 |
-
"loss": 0.
|
| 797 |
"step": 2700
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 4.99,
|
| 801 |
-
"grad_norm": 0.
|
| 802 |
-
"learning_rate": 0.
|
| 803 |
-
"loss": 0.
|
| 804 |
"step": 2725
|
| 805 |
},
|
| 806 |
{
|
| 807 |
"epoch": 5.0,
|
| 808 |
-
"eval_loss": 0.
|
| 809 |
-
"eval_runtime":
|
| 810 |
-
"eval_samples_per_second": 1.
|
| 811 |
-
"eval_steps_per_second": 0.
|
| 812 |
"step": 2730
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 5.04,
|
| 816 |
-
"grad_norm": 0.
|
| 817 |
-
"learning_rate": 0.
|
| 818 |
-
"loss": 0.
|
| 819 |
"step": 2750
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 5.08,
|
| 823 |
-
"grad_norm": 0.
|
| 824 |
-
"learning_rate": 0.
|
| 825 |
-
"loss": 0.
|
| 826 |
"step": 2775
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 5.13,
|
| 830 |
-
"grad_norm": 0.
|
| 831 |
-
"learning_rate": 0.
|
| 832 |
-
"loss": 0.
|
| 833 |
"step": 2800
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 5.17,
|
| 837 |
-
"grad_norm": 0.
|
| 838 |
-
"learning_rate": 0.
|
| 839 |
-
"loss": 0.
|
| 840 |
"step": 2825
|
| 841 |
},
|
| 842 |
{
|
| 843 |
"epoch": 5.22,
|
| 844 |
-
"grad_norm": 0.
|
| 845 |
-
"learning_rate": 0.
|
| 846 |
-
"loss": 0.
|
| 847 |
"step": 2850
|
| 848 |
},
|
| 849 |
{
|
| 850 |
"epoch": 5.27,
|
| 851 |
-
"grad_norm": 0.
|
| 852 |
-
"learning_rate": 0.
|
| 853 |
-
"loss": 0.
|
| 854 |
"step": 2875
|
| 855 |
},
|
| 856 |
{
|
| 857 |
"epoch": 5.31,
|
| 858 |
-
"grad_norm": 0.
|
| 859 |
-
"learning_rate": 0.
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 2900
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 5.36,
|
| 865 |
-
"grad_norm": 0.
|
| 866 |
-
"learning_rate": 0.
|
| 867 |
-
"loss": 0.
|
| 868 |
"step": 2925
|
| 869 |
},
|
| 870 |
{
|
| 871 |
"epoch": 5.4,
|
| 872 |
-
"grad_norm": 0.
|
| 873 |
-
"learning_rate": 0.
|
| 874 |
-
"loss": 0.
|
| 875 |
"step": 2950
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 5.45,
|
| 879 |
-
"grad_norm": 0.
|
| 880 |
-
"learning_rate": 0.
|
| 881 |
-
"loss": 0.
|
| 882 |
"step": 2975
|
| 883 |
},
|
| 884 |
{
|
| 885 |
"epoch": 5.49,
|
| 886 |
-
"grad_norm": 0.
|
| 887 |
-
"learning_rate": 0.
|
| 888 |
-
"loss": 0.
|
| 889 |
"step": 3000
|
| 890 |
},
|
| 891 |
{
|
| 892 |
"epoch": 5.54,
|
| 893 |
-
"grad_norm": 0.
|
| 894 |
-
"learning_rate": 0.
|
| 895 |
-
"loss": 0.
|
| 896 |
"step": 3025
|
| 897 |
},
|
| 898 |
{
|
| 899 |
"epoch": 5.59,
|
| 900 |
-
"grad_norm": 0.
|
| 901 |
-
"learning_rate": 0.
|
| 902 |
-
"loss": 0.
|
| 903 |
"step": 3050
|
| 904 |
},
|
| 905 |
{
|
| 906 |
"epoch": 5.63,
|
| 907 |
-
"grad_norm": 0.
|
| 908 |
-
"learning_rate": 0.
|
| 909 |
-
"loss": 0.
|
| 910 |
"step": 3075
|
| 911 |
},
|
| 912 |
{
|
| 913 |
"epoch": 5.68,
|
| 914 |
-
"grad_norm": 0.
|
| 915 |
-
"learning_rate": 0.
|
| 916 |
-
"loss": 0.
|
| 917 |
"step": 3100
|
| 918 |
},
|
| 919 |
{
|
| 920 |
"epoch": 5.72,
|
| 921 |
-
"grad_norm": 0.
|
| 922 |
-
"learning_rate": 0.
|
| 923 |
-
"loss": 0.
|
| 924 |
"step": 3125
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 5.77,
|
| 928 |
-
"grad_norm": 0.
|
| 929 |
-
"learning_rate": 0.
|
| 930 |
-
"loss": 0.
|
| 931 |
"step": 3150
|
| 932 |
},
|
| 933 |
{
|
| 934 |
"epoch": 5.82,
|
| 935 |
-
"grad_norm": 0.
|
| 936 |
-
"learning_rate": 0.
|
| 937 |
-
"loss": 0.
|
| 938 |
"step": 3175
|
| 939 |
},
|
| 940 |
{
|
| 941 |
"epoch": 5.86,
|
| 942 |
-
"grad_norm": 0.
|
| 943 |
-
"learning_rate": 0.
|
| 944 |
-
"loss": 0.
|
| 945 |
"step": 3200
|
| 946 |
},
|
| 947 |
{
|
| 948 |
"epoch": 5.91,
|
| 949 |
-
"grad_norm": 0.
|
| 950 |
-
"learning_rate": 0.
|
| 951 |
-
"loss": 0.
|
| 952 |
"step": 3225
|
| 953 |
},
|
| 954 |
{
|
| 955 |
"epoch": 5.95,
|
| 956 |
-
"grad_norm": 0.
|
| 957 |
-
"learning_rate": 0.
|
| 958 |
-
"loss": 0.
|
| 959 |
"step": 3250
|
| 960 |
},
|
| 961 |
{
|
| 962 |
"epoch": 6.0,
|
| 963 |
-
"grad_norm": 0.
|
| 964 |
-
"learning_rate": 0.
|
| 965 |
-
"loss": 0.
|
| 966 |
"step": 3275
|
| 967 |
},
|
| 968 |
{
|
| 969 |
"epoch": 6.0,
|
| 970 |
-
"eval_loss": 0.
|
| 971 |
-
"eval_runtime": 463.
|
| 972 |
-
"eval_samples_per_second": 1.
|
| 973 |
-
"eval_steps_per_second": 0.
|
| 974 |
"step": 3276
|
| 975 |
},
|
| 976 |
{
|
| 977 |
"epoch": 6.04,
|
| 978 |
-
"grad_norm": 0.
|
| 979 |
-
"learning_rate": 0.
|
| 980 |
-
"loss": 0.
|
| 981 |
"step": 3300
|
| 982 |
},
|
| 983 |
{
|
| 984 |
"epoch": 6.09,
|
| 985 |
-
"grad_norm": 0.
|
| 986 |
-
"learning_rate": 0.
|
| 987 |
-
"loss": 0.
|
| 988 |
"step": 3325
|
| 989 |
},
|
| 990 |
{
|
| 991 |
"epoch": 6.14,
|
| 992 |
-
"grad_norm": 0.
|
| 993 |
-
"learning_rate": 0.
|
| 994 |
-
"loss": 0.
|
| 995 |
"step": 3350
|
| 996 |
},
|
| 997 |
{
|
| 998 |
"epoch": 6.18,
|
| 999 |
-
"grad_norm": 0.
|
| 1000 |
-
"learning_rate": 0.
|
| 1001 |
-
"loss": 0.
|
| 1002 |
"step": 3375
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
"epoch": 6.23,
|
| 1006 |
-
"grad_norm": 0.
|
| 1007 |
-
"learning_rate": 0.
|
| 1008 |
-
"loss": 0.
|
| 1009 |
"step": 3400
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
"epoch": 6.27,
|
| 1013 |
-
"grad_norm": 0.
|
| 1014 |
-
"learning_rate": 0.
|
| 1015 |
-
"loss": 0.
|
| 1016 |
"step": 3425
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
"epoch": 6.32,
|
| 1020 |
-
"grad_norm": 0.
|
| 1021 |
-
"learning_rate": 0.
|
| 1022 |
-
"loss": 0.
|
| 1023 |
"step": 3450
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
"epoch": 6.36,
|
| 1027 |
-
"grad_norm": 0.
|
| 1028 |
-
"learning_rate": 0.
|
| 1029 |
-
"loss": 0.
|
| 1030 |
"step": 3475
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
"epoch": 6.41,
|
| 1034 |
-
"grad_norm": 0.
|
| 1035 |
-
"learning_rate": 0.
|
| 1036 |
-
"loss": 0.
|
| 1037 |
"step": 3500
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
"epoch": 6.46,
|
| 1041 |
-
"grad_norm": 0.
|
| 1042 |
-
"learning_rate": 0.
|
| 1043 |
-
"loss": 0.
|
| 1044 |
"step": 3525
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
"epoch": 6.5,
|
| 1048 |
-
"grad_norm": 0.
|
| 1049 |
-
"learning_rate": 0.
|
| 1050 |
-
"loss": 0.
|
| 1051 |
"step": 3550
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
"epoch": 6.55,
|
| 1055 |
-
"grad_norm": 0.
|
| 1056 |
-
"learning_rate": 0.
|
| 1057 |
-
"loss": 0.
|
| 1058 |
"step": 3575
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
"epoch": 6.59,
|
| 1062 |
-
"grad_norm": 0.
|
| 1063 |
-
"learning_rate": 0.
|
| 1064 |
-
"loss": 0.
|
| 1065 |
"step": 3600
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
"epoch": 6.64,
|
| 1069 |
-
"grad_norm": 0.
|
| 1070 |
-
"learning_rate": 0.
|
| 1071 |
-
"loss": 0.
|
| 1072 |
"step": 3625
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
"epoch": 6.68,
|
| 1076 |
-
"grad_norm": 0.
|
| 1077 |
-
"learning_rate": 0.
|
| 1078 |
-
"loss": 0.
|
| 1079 |
"step": 3650
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
"epoch": 6.73,
|
| 1083 |
-
"grad_norm": 0.
|
| 1084 |
-
"learning_rate": 0.
|
| 1085 |
-
"loss": 0.
|
| 1086 |
"step": 3675
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
"epoch": 6.78,
|
| 1090 |
-
"grad_norm": 0.
|
| 1091 |
-
"learning_rate": 0.
|
| 1092 |
-
"loss": 0.
|
| 1093 |
"step": 3700
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
"epoch": 6.82,
|
| 1097 |
-
"grad_norm": 0.
|
| 1098 |
-
"learning_rate": 0.
|
| 1099 |
-
"loss": 0.
|
| 1100 |
"step": 3725
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
"epoch": 6.87,
|
| 1104 |
-
"grad_norm": 0.
|
| 1105 |
-
"learning_rate": 0.
|
| 1106 |
-
"loss": 0.
|
| 1107 |
"step": 3750
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
"epoch": 6.91,
|
| 1111 |
-
"grad_norm": 0.
|
| 1112 |
-
"learning_rate": 0.
|
| 1113 |
-
"loss": 0.
|
| 1114 |
"step": 3775
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
"epoch": 6.96,
|
| 1118 |
-
"grad_norm": 0.
|
| 1119 |
-
"learning_rate": 0.
|
| 1120 |
-
"loss": 0.
|
| 1121 |
"step": 3800
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
"epoch": 7.0,
|
| 1125 |
-
"eval_loss": 0.
|
| 1126 |
-
"eval_runtime":
|
| 1127 |
-
"eval_samples_per_second": 1.
|
| 1128 |
-
"eval_steps_per_second": 0.
|
| 1129 |
"step": 3822
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 7.01,
|
| 1133 |
-
"grad_norm": 0.
|
| 1134 |
-
"learning_rate": 0.
|
| 1135 |
-
"loss": 0.
|
| 1136 |
"step": 3825
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
"epoch": 7.05,
|
| 1140 |
-
"grad_norm": 0.
|
| 1141 |
-
"learning_rate": 0.
|
| 1142 |
-
"loss": 0.
|
| 1143 |
"step": 3850
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
"epoch": 7.1,
|
| 1147 |
-
"grad_norm": 0.
|
| 1148 |
-
"learning_rate": 0.
|
| 1149 |
-
"loss": 0.
|
| 1150 |
"step": 3875
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
"epoch": 7.14,
|
| 1154 |
-
"grad_norm": 0.
|
| 1155 |
-
"learning_rate": 0.
|
| 1156 |
-
"loss": 0.
|
| 1157 |
"step": 3900
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 7.19,
|
| 1161 |
-
"grad_norm": 0.
|
| 1162 |
-
"learning_rate": 0.
|
| 1163 |
-
"loss": 0.
|
| 1164 |
"step": 3925
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
"epoch": 7.23,
|
| 1168 |
-
"grad_norm": 0.
|
| 1169 |
-
"learning_rate": 0.
|
| 1170 |
-
"loss": 0.
|
| 1171 |
"step": 3950
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 7.28,
|
| 1175 |
-
"grad_norm": 0.
|
| 1176 |
-
"learning_rate": 0.
|
| 1177 |
-
"loss": 0.
|
| 1178 |
"step": 3975
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 7.33,
|
| 1182 |
-
"grad_norm": 0.
|
| 1183 |
-
"learning_rate": 0.
|
| 1184 |
-
"loss": 0.
|
| 1185 |
"step": 4000
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 7.37,
|
| 1189 |
-
"grad_norm": 0.
|
| 1190 |
-
"learning_rate": 0.
|
| 1191 |
-
"loss": 0.
|
| 1192 |
"step": 4025
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
"epoch": 7.42,
|
| 1196 |
-
"grad_norm": 0.
|
| 1197 |
-
"learning_rate": 0.
|
| 1198 |
-
"loss": 0.
|
| 1199 |
"step": 4050
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 7.46,
|
| 1203 |
-
"grad_norm": 0.
|
| 1204 |
-
"learning_rate": 0.
|
| 1205 |
-
"loss": 0.
|
| 1206 |
"step": 4075
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
"epoch": 7.51,
|
| 1210 |
-
"grad_norm": 0.
|
| 1211 |
-
"learning_rate": 0.
|
| 1212 |
-
"loss": 0.
|
| 1213 |
"step": 4100
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 7.55,
|
| 1217 |
-
"grad_norm": 0.
|
| 1218 |
-
"learning_rate": 0.
|
| 1219 |
-
"loss": 0.
|
| 1220 |
"step": 4125
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
"epoch": 7.6,
|
| 1224 |
-
"grad_norm": 0.
|
| 1225 |
-
"learning_rate": 0.
|
| 1226 |
-
"loss": 0.
|
| 1227 |
"step": 4150
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 7.65,
|
| 1231 |
-
"grad_norm": 0.
|
| 1232 |
-
"learning_rate": 0.
|
| 1233 |
-
"loss": 0.
|
| 1234 |
"step": 4175
|
| 1235 |
},
|
| 1236 |
{
|
| 1237 |
"epoch": 7.69,
|
| 1238 |
-
"grad_norm": 0.
|
| 1239 |
-
"learning_rate": 0.
|
| 1240 |
-
"loss": 0.
|
| 1241 |
"step": 4200
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 7.74,
|
| 1245 |
-
"grad_norm": 0.
|
| 1246 |
-
"learning_rate": 0.
|
| 1247 |
-
"loss": 0.
|
| 1248 |
"step": 4225
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
"epoch": 7.78,
|
| 1252 |
-
"grad_norm": 0.
|
| 1253 |
-
"learning_rate": 0.
|
| 1254 |
-
"loss": 0.
|
| 1255 |
"step": 4250
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
"epoch": 7.83,
|
| 1259 |
-
"grad_norm": 0.
|
| 1260 |
-
"learning_rate": 0.
|
| 1261 |
-
"loss": 0.
|
| 1262 |
"step": 4275
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
"epoch": 7.88,
|
| 1266 |
-
"grad_norm": 0.
|
| 1267 |
-
"learning_rate": 0.
|
| 1268 |
-
"loss": 0.
|
| 1269 |
"step": 4300
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 7.92,
|
| 1273 |
-
"grad_norm": 0.
|
| 1274 |
-
"learning_rate": 0.
|
| 1275 |
-
"loss": 0.
|
| 1276 |
"step": 4325
|
| 1277 |
},
|
| 1278 |
{
|
| 1279 |
"epoch": 7.97,
|
| 1280 |
-
"grad_norm": 0.
|
| 1281 |
-
"learning_rate": 0.
|
| 1282 |
-
"loss": 0.
|
| 1283 |
"step": 4350
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
"epoch": 8.0,
|
| 1287 |
-
"eval_loss": 0.
|
| 1288 |
-
"eval_runtime":
|
| 1289 |
-
"eval_samples_per_second": 1.
|
| 1290 |
-
"eval_steps_per_second": 0.
|
| 1291 |
"step": 4368
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
"epoch": 8.01,
|
| 1295 |
-
"grad_norm": 0.
|
| 1296 |
-
"learning_rate": 0.
|
| 1297 |
-
"loss": 0.
|
| 1298 |
"step": 4375
|
| 1299 |
},
|
| 1300 |
{
|
| 1301 |
"epoch": 8.06,
|
| 1302 |
-
"grad_norm": 0.
|
| 1303 |
-
"learning_rate": 0.
|
| 1304 |
-
"loss": 0.
|
| 1305 |
"step": 4400
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
"epoch": 8.1,
|
| 1309 |
-
"grad_norm": 0.
|
| 1310 |
-
"learning_rate": 0.
|
| 1311 |
-
"loss": 0.
|
| 1312 |
"step": 4425
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
"epoch": 8.15,
|
| 1316 |
-
"grad_norm": 0.
|
| 1317 |
-
"learning_rate": 0.
|
| 1318 |
-
"loss": 0.
|
| 1319 |
"step": 4450
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
"epoch": 8.2,
|
| 1323 |
-
"grad_norm": 0.
|
| 1324 |
-
"learning_rate": 0.
|
| 1325 |
-
"loss": 0.
|
| 1326 |
"step": 4475
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
"epoch": 8.24,
|
| 1330 |
-
"grad_norm": 0.
|
| 1331 |
-
"learning_rate": 0.
|
| 1332 |
-
"loss": 0.
|
| 1333 |
"step": 4500
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
"epoch": 8.29,
|
| 1337 |
-
"grad_norm": 0.
|
| 1338 |
-
"learning_rate": 0.
|
| 1339 |
-
"loss": 0.
|
| 1340 |
"step": 4525
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
"epoch": 8.33,
|
| 1344 |
-
"grad_norm": 0.
|
| 1345 |
-
"learning_rate": 0.
|
| 1346 |
-
"loss": 0.
|
| 1347 |
"step": 4550
|
| 1348 |
},
|
| 1349 |
{
|
| 1350 |
"epoch": 8.38,
|
| 1351 |
-
"grad_norm": 0.
|
| 1352 |
-
"learning_rate": 0.
|
| 1353 |
-
"loss": 0.
|
| 1354 |
"step": 4575
|
| 1355 |
},
|
| 1356 |
{
|
| 1357 |
"epoch": 8.42,
|
| 1358 |
-
"grad_norm": 0.
|
| 1359 |
-
"learning_rate": 0.
|
| 1360 |
-
"loss": 0.
|
| 1361 |
"step": 4600
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
"epoch": 8.47,
|
| 1365 |
-
"grad_norm": 0.
|
| 1366 |
-
"learning_rate": 0.
|
| 1367 |
-
"loss": 0.
|
| 1368 |
"step": 4625
|
| 1369 |
},
|
| 1370 |
{
|
| 1371 |
"epoch": 8.52,
|
| 1372 |
-
"grad_norm": 0.
|
| 1373 |
-
"learning_rate": 0.
|
| 1374 |
-
"loss": 0.
|
| 1375 |
"step": 4650
|
| 1376 |
},
|
| 1377 |
{
|
| 1378 |
"epoch": 8.56,
|
| 1379 |
-
"grad_norm": 0.
|
| 1380 |
-
"learning_rate": 0.
|
| 1381 |
-
"loss": 0.
|
| 1382 |
"step": 4675
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
"epoch": 8.61,
|
| 1386 |
-
"grad_norm": 0.
|
| 1387 |
-
"learning_rate": 0.
|
| 1388 |
-
"loss": 0.
|
| 1389 |
"step": 4700
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
"epoch": 8.65,
|
| 1393 |
-
"grad_norm": 0.
|
| 1394 |
-
"learning_rate": 0.
|
| 1395 |
-
"loss": 0.
|
| 1396 |
"step": 4725
|
| 1397 |
},
|
| 1398 |
{
|
| 1399 |
"epoch": 8.7,
|
| 1400 |
-
"grad_norm": 0.
|
| 1401 |
-
"learning_rate": 0.
|
| 1402 |
-
"loss": 0.
|
| 1403 |
"step": 4750
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
"epoch": 8.75,
|
| 1407 |
-
"grad_norm": 0.
|
| 1408 |
-
"learning_rate": 0.
|
| 1409 |
-
"loss": 0.
|
| 1410 |
"step": 4775
|
| 1411 |
},
|
| 1412 |
{
|
| 1413 |
"epoch": 8.79,
|
| 1414 |
-
"grad_norm": 0.
|
| 1415 |
-
"learning_rate": 0.
|
| 1416 |
-
"loss": 0.
|
| 1417 |
"step": 4800
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
"epoch": 8.84,
|
| 1421 |
-
"grad_norm": 0.
|
| 1422 |
-
"learning_rate": 0.
|
| 1423 |
-
"loss": 0.
|
| 1424 |
"step": 4825
|
| 1425 |
},
|
| 1426 |
{
|
| 1427 |
"epoch": 8.88,
|
| 1428 |
-
"grad_norm": 0.
|
| 1429 |
-
"learning_rate": 0.
|
| 1430 |
-
"loss": 0.
|
| 1431 |
"step": 4850
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
"epoch": 8.93,
|
| 1435 |
-
"grad_norm": 0.
|
| 1436 |
-
"learning_rate": 0.
|
| 1437 |
-
"loss": 0.
|
| 1438 |
"step": 4875
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
"epoch": 8.97,
|
| 1442 |
-
"grad_norm": 0.
|
| 1443 |
-
"learning_rate": 0.
|
| 1444 |
-
"loss": 0.
|
| 1445 |
"step": 4900
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
"epoch": 9.0,
|
| 1449 |
-
"eval_loss": 0.
|
| 1450 |
-
"eval_runtime":
|
| 1451 |
-
"eval_samples_per_second": 1.
|
| 1452 |
-
"eval_steps_per_second": 0.
|
| 1453 |
"step": 4914
|
| 1454 |
},
|
| 1455 |
{
|
| 1456 |
"epoch": 9.02,
|
| 1457 |
-
"grad_norm": 0.
|
| 1458 |
-
"learning_rate": 9.
|
| 1459 |
-
"loss": 0.
|
| 1460 |
"step": 4925
|
| 1461 |
},
|
| 1462 |
{
|
| 1463 |
"epoch": 9.07,
|
| 1464 |
-
"grad_norm": 0.
|
| 1465 |
-
"learning_rate": 9.
|
| 1466 |
-
"loss": 0.
|
| 1467 |
"step": 4950
|
| 1468 |
},
|
| 1469 |
{
|
| 1470 |
"epoch": 9.11,
|
| 1471 |
-
"grad_norm": 0.
|
| 1472 |
-
"learning_rate":
|
| 1473 |
-
"loss": 0.
|
| 1474 |
"step": 4975
|
| 1475 |
},
|
| 1476 |
{
|
| 1477 |
"epoch": 9.16,
|
| 1478 |
-
"grad_norm": 0.
|
| 1479 |
-
"learning_rate": 8.
|
| 1480 |
-
"loss": 0.
|
| 1481 |
"step": 5000
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
"epoch": 9.2,
|
| 1485 |
-
"grad_norm": 0.
|
| 1486 |
-
"learning_rate": 8.
|
| 1487 |
-
"loss": 0.
|
| 1488 |
"step": 5025
|
| 1489 |
},
|
| 1490 |
{
|
| 1491 |
"epoch": 9.25,
|
| 1492 |
-
"grad_norm": 0.
|
| 1493 |
-
"learning_rate": 7.
|
| 1494 |
-
"loss": 0.
|
| 1495 |
"step": 5050
|
| 1496 |
},
|
| 1497 |
{
|
| 1498 |
"epoch": 9.29,
|
| 1499 |
-
"grad_norm": 0.
|
| 1500 |
-
"learning_rate": 7.
|
| 1501 |
-
"loss": 0.
|
| 1502 |
"step": 5075
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
"epoch": 9.34,
|
| 1506 |
-
"grad_norm": 0.
|
| 1507 |
-
"learning_rate": 6.
|
| 1508 |
-
"loss": 0.
|
| 1509 |
"step": 5100
|
| 1510 |
},
|
| 1511 |
{
|
| 1512 |
"epoch": 9.39,
|
| 1513 |
-
"grad_norm": 0.
|
| 1514 |
-
"learning_rate": 6.
|
| 1515 |
-
"loss": 0.
|
| 1516 |
"step": 5125
|
| 1517 |
},
|
| 1518 |
{
|
| 1519 |
"epoch": 9.43,
|
| 1520 |
-
"grad_norm": 0.
|
| 1521 |
-
"learning_rate": 5.
|
| 1522 |
-
"loss": 0.
|
| 1523 |
"step": 5150
|
| 1524 |
},
|
| 1525 |
{
|
| 1526 |
"epoch": 9.48,
|
| 1527 |
-
"grad_norm": 0.
|
| 1528 |
-
"learning_rate": 5.
|
| 1529 |
-
"loss": 0.
|
| 1530 |
"step": 5175
|
| 1531 |
},
|
| 1532 |
{
|
| 1533 |
"epoch": 9.52,
|
| 1534 |
-
"grad_norm": 0.
|
| 1535 |
-
"learning_rate": 4.
|
| 1536 |
-
"loss": 0.
|
| 1537 |
"step": 5200
|
| 1538 |
},
|
| 1539 |
{
|
| 1540 |
"epoch": 9.57,
|
| 1541 |
-
"grad_norm": 0.
|
| 1542 |
-
"learning_rate": 4.
|
| 1543 |
-
"loss": 0.
|
| 1544 |
"step": 5225
|
| 1545 |
},
|
| 1546 |
{
|
| 1547 |
"epoch": 9.62,
|
| 1548 |
-
"grad_norm": 0.
|
| 1549 |
-
"learning_rate": 3.
|
| 1550 |
-
"loss": 0.
|
| 1551 |
"step": 5250
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
"epoch": 9.66,
|
| 1555 |
-
"grad_norm": 0.
|
| 1556 |
-
"learning_rate": 3.
|
| 1557 |
-
"loss": 0.
|
| 1558 |
"step": 5275
|
| 1559 |
},
|
| 1560 |
{
|
| 1561 |
"epoch": 9.71,
|
| 1562 |
-
"grad_norm": 0.
|
| 1563 |
-
"learning_rate":
|
| 1564 |
-
"loss": 0.
|
| 1565 |
"step": 5300
|
| 1566 |
},
|
| 1567 |
{
|
| 1568 |
"epoch": 9.75,
|
| 1569 |
-
"grad_norm": 0.
|
| 1570 |
-
"learning_rate": 2.
|
| 1571 |
-
"loss": 0.
|
| 1572 |
"step": 5325
|
| 1573 |
},
|
| 1574 |
{
|
| 1575 |
"epoch": 9.8,
|
| 1576 |
-
"grad_norm": 0.
|
| 1577 |
-
"learning_rate": 2.
|
| 1578 |
-
"loss": 0.
|
| 1579 |
"step": 5350
|
| 1580 |
},
|
| 1581 |
{
|
| 1582 |
"epoch": 9.84,
|
| 1583 |
-
"grad_norm": 0.
|
| 1584 |
-
"learning_rate": 1.
|
| 1585 |
-
"loss": 0.
|
| 1586 |
"step": 5375
|
| 1587 |
},
|
| 1588 |
{
|
| 1589 |
"epoch": 9.89,
|
| 1590 |
-
"grad_norm": 0.
|
| 1591 |
-
"learning_rate": 1.
|
| 1592 |
-
"loss": 0.
|
| 1593 |
"step": 5400
|
| 1594 |
},
|
| 1595 |
{
|
| 1596 |
"epoch": 9.94,
|
| 1597 |
-
"grad_norm": 0.
|
| 1598 |
-
"learning_rate":
|
| 1599 |
-
"loss": 0.
|
| 1600 |
"step": 5425
|
| 1601 |
},
|
| 1602 |
{
|
| 1603 |
"epoch": 9.98,
|
| 1604 |
-
"grad_norm": 0.
|
| 1605 |
-
"learning_rate":
|
| 1606 |
-
"loss": 0.
|
| 1607 |
"step": 5450
|
| 1608 |
},
|
| 1609 |
{
|
| 1610 |
"epoch": 10.0,
|
| 1611 |
-
"eval_loss": 0.
|
| 1612 |
-
"eval_runtime":
|
| 1613 |
-
"eval_samples_per_second": 1.
|
| 1614 |
-
"eval_steps_per_second": 0.
|
| 1615 |
"step": 5460
|
| 1616 |
},
|
| 1617 |
{
|
| 1618 |
"epoch": 10.0,
|
| 1619 |
"step": 5460,
|
| 1620 |
"total_flos": 1.135723105419264e+20,
|
| 1621 |
-
"train_loss": 0.
|
| 1622 |
-
"train_runtime":
|
| 1623 |
-
"train_samples_per_second": 1.
|
| 1624 |
"train_steps_per_second": 0.188
|
| 1625 |
}
|
| 1626 |
],
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05,
|
| 13 |
+
"grad_norm": 0.0019010665128007531,
|
| 14 |
"learning_rate": 0.0005,
|
| 15 |
+
"loss": 0.0009,
|
| 16 |
"step": 25
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09,
|
| 20 |
+
"grad_norm": 0.33505979180336,
|
| 21 |
"learning_rate": 0.001,
|
| 22 |
+
"loss": 0.0059,
|
| 23 |
"step": 50
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.14,
|
| 27 |
+
"grad_norm": 2.521899700164795,
|
| 28 |
+
"learning_rate": 0.0009953789279112755,
|
| 29 |
+
"loss": 0.2569,
|
| 30 |
"step": 75
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.18,
|
| 34 |
+
"grad_norm": 1.3200310468673706,
|
| 35 |
+
"learning_rate": 0.000990757855822551,
|
| 36 |
+
"loss": 2.1666,
|
| 37 |
"step": 100
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.23,
|
| 41 |
+
"grad_norm": 1.1342906951904297,
|
| 42 |
+
"learning_rate": 0.0009861367837338264,
|
| 43 |
+
"loss": 0.3677,
|
| 44 |
"step": 125
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.27,
|
| 48 |
+
"grad_norm": 0.6839067935943604,
|
| 49 |
+
"learning_rate": 0.0009815157116451016,
|
| 50 |
+
"loss": 0.2573,
|
| 51 |
"step": 150
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.32,
|
| 55 |
+
"grad_norm": 1.120778203010559,
|
| 56 |
+
"learning_rate": 0.0009768946395563771,
|
| 57 |
+
"loss": 0.1784,
|
| 58 |
"step": 175
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.37,
|
| 62 |
+
"grad_norm": 0.39630964398384094,
|
| 63 |
+
"learning_rate": 0.0009722735674676525,
|
| 64 |
+
"loss": 0.1007,
|
| 65 |
"step": 200
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.41,
|
| 69 |
+
"grad_norm": 0.8115565776824951,
|
| 70 |
+
"learning_rate": 0.0009676524953789279,
|
| 71 |
+
"loss": 0.1009,
|
| 72 |
"step": 225
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.46,
|
| 76 |
+
"grad_norm": 0.4256753921508789,
|
| 77 |
+
"learning_rate": 0.0009630314232902033,
|
| 78 |
+
"loss": 0.0613,
|
| 79 |
"step": 250
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.5,
|
| 83 |
+
"grad_norm": 1.3296879529953003,
|
| 84 |
+
"learning_rate": 0.0009584103512014787,
|
| 85 |
+
"loss": 0.0656,
|
| 86 |
"step": 275
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.55,
|
| 90 |
+
"grad_norm": 0.2332513928413391,
|
| 91 |
+
"learning_rate": 0.0009537892791127542,
|
| 92 |
+
"loss": 0.1056,
|
| 93 |
"step": 300
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.6,
|
| 97 |
+
"grad_norm": 0.361102819442749,
|
| 98 |
+
"learning_rate": 0.0009491682070240297,
|
| 99 |
+
"loss": 0.0714,
|
| 100 |
"step": 325
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.64,
|
| 104 |
+
"grad_norm": 0.3176443576812744,
|
| 105 |
+
"learning_rate": 0.000944547134935305,
|
| 106 |
+
"loss": 0.051,
|
| 107 |
"step": 350
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.69,
|
| 111 |
+
"grad_norm": 0.5561681389808655,
|
| 112 |
+
"learning_rate": 0.0009399260628465805,
|
| 113 |
+
"loss": 0.0987,
|
| 114 |
"step": 375
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.73,
|
| 118 |
+
"grad_norm": 0.17937970161437988,
|
| 119 |
+
"learning_rate": 0.0009353049907578558,
|
| 120 |
+
"loss": 0.0705,
|
| 121 |
"step": 400
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.78,
|
| 125 |
+
"grad_norm": 0.11134567111730576,
|
| 126 |
+
"learning_rate": 0.0009306839186691313,
|
| 127 |
+
"loss": 0.0449,
|
| 128 |
"step": 425
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.82,
|
| 132 |
+
"grad_norm": 0.1100844293832779,
|
| 133 |
+
"learning_rate": 0.0009260628465804066,
|
| 134 |
+
"loss": 0.0605,
|
| 135 |
"step": 450
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.87,
|
| 139 |
+
"grad_norm": 0.5044831037521362,
|
| 140 |
+
"learning_rate": 0.0009214417744916821,
|
| 141 |
+
"loss": 0.0682,
|
| 142 |
"step": 475
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.92,
|
| 146 |
+
"grad_norm": 0.1506507396697998,
|
| 147 |
+
"learning_rate": 0.0009168207024029575,
|
| 148 |
+
"loss": 0.0382,
|
| 149 |
"step": 500
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.96,
|
| 153 |
+
"grad_norm": 0.358964741230011,
|
| 154 |
+
"learning_rate": 0.0009121996303142329,
|
| 155 |
+
"loss": 0.0416,
|
| 156 |
"step": 525
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 1.0,
|
| 160 |
+
"eval_loss": 0.22685140371322632,
|
| 161 |
+
"eval_runtime": 467.2357,
|
| 162 |
+
"eval_samples_per_second": 1.74,
|
| 163 |
+
"eval_steps_per_second": 0.291,
|
| 164 |
"step": 546
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 1.01,
|
| 168 |
+
"grad_norm": 0.21843843162059784,
|
| 169 |
+
"learning_rate": 0.0009075785582255084,
|
| 170 |
+
"loss": 0.0283,
|
| 171 |
"step": 550
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 1.05,
|
| 175 |
+
"grad_norm": 0.3907661437988281,
|
| 176 |
+
"learning_rate": 0.0009029574861367837,
|
| 177 |
+
"loss": 0.0322,
|
| 178 |
"step": 575
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 1.1,
|
| 182 |
+
"grad_norm": 0.026330502703785896,
|
| 183 |
+
"learning_rate": 0.0008983364140480592,
|
| 184 |
+
"loss": 0.021,
|
| 185 |
"step": 600
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 1.14,
|
| 189 |
+
"grad_norm": 0.015783503651618958,
|
| 190 |
+
"learning_rate": 0.0008937153419593346,
|
| 191 |
+
"loss": 0.0237,
|
| 192 |
"step": 625
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 1.19,
|
| 196 |
+
"grad_norm": 0.07014349848031998,
|
| 197 |
+
"learning_rate": 0.00088909426987061,
|
| 198 |
+
"loss": 0.0307,
|
| 199 |
"step": 650
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 1.24,
|
| 203 |
+
"grad_norm": 0.05321989953517914,
|
| 204 |
+
"learning_rate": 0.0008844731977818854,
|
| 205 |
+
"loss": 0.0217,
|
| 206 |
"step": 675
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 1.28,
|
| 210 |
+
"grad_norm": 0.07509706914424896,
|
| 211 |
+
"learning_rate": 0.0008798521256931608,
|
| 212 |
+
"loss": 0.0319,
|
| 213 |
"step": 700
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 1.33,
|
| 217 |
+
"grad_norm": 0.239598348736763,
|
| 218 |
+
"learning_rate": 0.0008752310536044362,
|
| 219 |
+
"loss": 0.0373,
|
| 220 |
"step": 725
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 1.37,
|
| 224 |
+
"grad_norm": 0.1276847869157791,
|
| 225 |
+
"learning_rate": 0.0008706099815157116,
|
| 226 |
+
"loss": 0.0308,
|
| 227 |
"step": 750
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 1.42,
|
| 231 |
+
"grad_norm": 0.17856040596961975,
|
| 232 |
+
"learning_rate": 0.000865988909426987,
|
| 233 |
+
"loss": 0.0326,
|
| 234 |
"step": 775
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 1.47,
|
| 238 |
+
"grad_norm": 0.5299984216690063,
|
| 239 |
+
"learning_rate": 0.0008613678373382626,
|
| 240 |
+
"loss": 0.034,
|
| 241 |
"step": 800
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 1.51,
|
| 245 |
+
"grad_norm": 0.18570055067539215,
|
| 246 |
+
"learning_rate": 0.0008567467652495379,
|
| 247 |
+
"loss": 0.0255,
|
| 248 |
"step": 825
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 1.56,
|
| 252 |
+
"grad_norm": 0.012400169856846333,
|
| 253 |
+
"learning_rate": 0.0008521256931608134,
|
| 254 |
+
"loss": 0.0133,
|
| 255 |
"step": 850
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 1.6,
|
| 259 |
+
"grad_norm": 0.18435439467430115,
|
| 260 |
+
"learning_rate": 0.0008475046210720887,
|
| 261 |
+
"loss": 0.0392,
|
| 262 |
"step": 875
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 1.65,
|
| 266 |
+
"grad_norm": 0.20227985084056854,
|
| 267 |
+
"learning_rate": 0.0008428835489833642,
|
| 268 |
+
"loss": 0.0339,
|
| 269 |
"step": 900
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 1.69,
|
| 273 |
+
"grad_norm": 0.09818145632743835,
|
| 274 |
+
"learning_rate": 0.0008382624768946395,
|
| 275 |
+
"loss": 0.039,
|
| 276 |
"step": 925
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 1.74,
|
| 280 |
+
"grad_norm": 0.17143449187278748,
|
| 281 |
+
"learning_rate": 0.000833641404805915,
|
| 282 |
+
"loss": 0.0256,
|
| 283 |
"step": 950
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 1.79,
|
| 287 |
+
"grad_norm": 0.3052353858947754,
|
| 288 |
+
"learning_rate": 0.0008290203327171904,
|
| 289 |
+
"loss": 0.0279,
|
| 290 |
"step": 975
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 1.83,
|
| 294 |
+
"grad_norm": 0.09069275110960007,
|
| 295 |
+
"learning_rate": 0.0008243992606284658,
|
| 296 |
+
"loss": 0.0253,
|
| 297 |
"step": 1000
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 1.88,
|
| 301 |
+
"grad_norm": 0.7721070647239685,
|
| 302 |
+
"learning_rate": 0.0008197781885397413,
|
| 303 |
+
"loss": 0.0222,
|
| 304 |
"step": 1025
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 1.92,
|
| 308 |
+
"grad_norm": 0.2378959357738495,
|
| 309 |
+
"learning_rate": 0.0008151571164510166,
|
| 310 |
+
"loss": 0.0386,
|
| 311 |
"step": 1050
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 1.97,
|
| 315 |
+
"grad_norm": 0.004904525820165873,
|
| 316 |
+
"learning_rate": 0.0008105360443622921,
|
| 317 |
+
"loss": 0.0243,
|
| 318 |
"step": 1075
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 2.0,
|
| 322 |
+
"eval_loss": 0.20544852316379547,
|
| 323 |
+
"eval_runtime": 463.5714,
|
| 324 |
+
"eval_samples_per_second": 1.754,
|
| 325 |
+
"eval_steps_per_second": 0.293,
|
| 326 |
"step": 1092
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 2.01,
|
| 330 |
+
"grad_norm": 0.13851934671401978,
|
| 331 |
+
"learning_rate": 0.0008059149722735675,
|
| 332 |
+
"loss": 0.0267,
|
| 333 |
"step": 1100
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 2.06,
|
| 337 |
+
"grad_norm": 0.01875193975865841,
|
| 338 |
+
"learning_rate": 0.0008012939001848429,
|
| 339 |
+
"loss": 0.0117,
|
| 340 |
"step": 1125
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 2.11,
|
| 344 |
+
"grad_norm": 0.0016854548593983054,
|
| 345 |
+
"learning_rate": 0.0007966728280961183,
|
| 346 |
+
"loss": 0.0317,
|
| 347 |
"step": 1150
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 2.15,
|
| 351 |
+
"grad_norm": 0.10259977728128433,
|
| 352 |
+
"learning_rate": 0.0007920517560073937,
|
| 353 |
+
"loss": 0.0253,
|
| 354 |
"step": 1175
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 2.2,
|
| 358 |
+
"grad_norm": 0.054936520755290985,
|
| 359 |
+
"learning_rate": 0.0007874306839186691,
|
| 360 |
+
"loss": 0.0228,
|
| 361 |
"step": 1200
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 2.24,
|
| 365 |
+
"grad_norm": 0.08183781057596207,
|
| 366 |
+
"learning_rate": 0.0007828096118299445,
|
| 367 |
+
"loss": 0.0163,
|
| 368 |
"step": 1225
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 2.29,
|
| 372 |
+
"grad_norm": 0.03448671102523804,
|
| 373 |
+
"learning_rate": 0.0007781885397412199,
|
| 374 |
+
"loss": 0.0252,
|
| 375 |
"step": 1250
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 2.34,
|
| 379 |
+
"grad_norm": 0.026254719123244286,
|
| 380 |
+
"learning_rate": 0.0007735674676524955,
|
| 381 |
+
"loss": 0.0263,
|
| 382 |
"step": 1275
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 2.38,
|
| 386 |
+
"grad_norm": 0.07833431661128998,
|
| 387 |
+
"learning_rate": 0.0007689463955637708,
|
| 388 |
+
"loss": 0.0268,
|
| 389 |
"step": 1300
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 2.43,
|
| 393 |
+
"grad_norm": 0.3215916156768799,
|
| 394 |
+
"learning_rate": 0.0007643253234750463,
|
| 395 |
+
"loss": 0.0184,
|
| 396 |
"step": 1325
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 2.47,
|
| 400 |
+
"grad_norm": 0.1949949562549591,
|
| 401 |
+
"learning_rate": 0.0007597042513863216,
|
| 402 |
+
"loss": 0.0234,
|
| 403 |
"step": 1350
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 2.52,
|
| 407 |
+
"grad_norm": 0.10761301964521408,
|
| 408 |
+
"learning_rate": 0.0007550831792975971,
|
| 409 |
+
"loss": 0.0214,
|
| 410 |
"step": 1375
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 2.56,
|
| 414 |
+
"grad_norm": 0.23488566279411316,
|
| 415 |
+
"learning_rate": 0.0007504621072088724,
|
| 416 |
+
"loss": 0.0321,
|
| 417 |
"step": 1400
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 2.61,
|
| 421 |
+
"grad_norm": 0.16079489886760712,
|
| 422 |
+
"learning_rate": 0.0007458410351201479,
|
| 423 |
+
"loss": 0.041,
|
| 424 |
"step": 1425
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 2.66,
|
| 428 |
+
"grad_norm": 0.3500367999076843,
|
| 429 |
+
"learning_rate": 0.0007412199630314234,
|
| 430 |
+
"loss": 0.0218,
|
| 431 |
"step": 1450
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 2.7,
|
| 435 |
+
"grad_norm": 0.01966880075633526,
|
| 436 |
+
"learning_rate": 0.0007365988909426987,
|
| 437 |
+
"loss": 0.0185,
|
| 438 |
"step": 1475
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 2.75,
|
| 442 |
+
"grad_norm": 0.10084854066371918,
|
| 443 |
+
"learning_rate": 0.0007319778188539742,
|
| 444 |
+
"loss": 0.0158,
|
| 445 |
"step": 1500
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 2.79,
|
| 449 |
+
"grad_norm": 0.045843809843063354,
|
| 450 |
+
"learning_rate": 0.0007273567467652495,
|
| 451 |
+
"loss": 0.0193,
|
| 452 |
"step": 1525
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 2.84,
|
| 456 |
+
"grad_norm": 0.19230197370052338,
|
| 457 |
+
"learning_rate": 0.000722735674676525,
|
| 458 |
+
"loss": 0.0115,
|
| 459 |
"step": 1550
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 2.88,
|
| 463 |
+
"grad_norm": 0.10168833285570145,
|
| 464 |
+
"learning_rate": 0.0007181146025878004,
|
| 465 |
+
"loss": 0.0173,
|
| 466 |
"step": 1575
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 2.93,
|
| 470 |
+
"grad_norm": 0.24770613014698029,
|
| 471 |
+
"learning_rate": 0.0007134935304990758,
|
| 472 |
+
"loss": 0.019,
|
| 473 |
"step": 1600
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 2.98,
|
| 477 |
+
"grad_norm": 0.04277370125055313,
|
| 478 |
+
"learning_rate": 0.0007088724584103512,
|
| 479 |
+
"loss": 0.0262,
|
| 480 |
"step": 1625
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 3.0,
|
| 484 |
+
"eval_loss": 0.18655328452587128,
|
| 485 |
+
"eval_runtime": 463.8687,
|
| 486 |
+
"eval_samples_per_second": 1.753,
|
| 487 |
+
"eval_steps_per_second": 0.293,
|
| 488 |
"step": 1638
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 3.02,
|
| 492 |
+
"grad_norm": 0.03143952414393425,
|
| 493 |
+
"learning_rate": 0.0007042513863216266,
|
| 494 |
+
"loss": 0.0185,
|
| 495 |
"step": 1650
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 3.07,
|
| 499 |
+
"grad_norm": 0.061480745673179626,
|
| 500 |
+
"learning_rate": 0.000699630314232902,
|
| 501 |
+
"loss": 0.0158,
|
| 502 |
"step": 1675
|
| 503 |
},
|
| 504 |
{
|
| 505 |
"epoch": 3.11,
|
| 506 |
+
"grad_norm": 0.05645143985748291,
|
| 507 |
+
"learning_rate": 0.0006950092421441774,
|
| 508 |
+
"loss": 0.0163,
|
| 509 |
"step": 1700
|
| 510 |
},
|
| 511 |
{
|
| 512 |
"epoch": 3.16,
|
| 513 |
+
"grad_norm": 0.3927539885044098,
|
| 514 |
+
"learning_rate": 0.0006903881700554528,
|
| 515 |
+
"loss": 0.0257,
|
| 516 |
"step": 1725
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 3.21,
|
| 520 |
+
"grad_norm": 0.1579461544752121,
|
| 521 |
+
"learning_rate": 0.0006857670979667284,
|
| 522 |
+
"loss": 0.0203,
|
| 523 |
"step": 1750
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 3.25,
|
| 527 |
+
"grad_norm": 0.003284105099737644,
|
| 528 |
+
"learning_rate": 0.0006811460258780037,
|
| 529 |
+
"loss": 0.012,
|
| 530 |
"step": 1775
|
| 531 |
},
|
| 532 |
{
|
| 533 |
"epoch": 3.3,
|
| 534 |
+
"grad_norm": 0.0939943715929985,
|
| 535 |
+
"learning_rate": 0.0006765249537892792,
|
| 536 |
+
"loss": 0.0139,
|
| 537 |
"step": 1800
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 3.34,
|
| 541 |
+
"grad_norm": 0.08114974200725555,
|
| 542 |
+
"learning_rate": 0.0006719038817005545,
|
| 543 |
+
"loss": 0.0134,
|
| 544 |
"step": 1825
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 3.39,
|
| 548 |
+
"grad_norm": 0.008277042768895626,
|
| 549 |
+
"learning_rate": 0.00066728280961183,
|
| 550 |
+
"loss": 0.0203,
|
| 551 |
"step": 1850
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 3.43,
|
| 555 |
+
"grad_norm": 0.014137201942503452,
|
| 556 |
+
"learning_rate": 0.0006626617375231053,
|
| 557 |
+
"loss": 0.018,
|
| 558 |
"step": 1875
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 3.48,
|
| 562 |
+
"grad_norm": 0.04209378361701965,
|
| 563 |
+
"learning_rate": 0.0006580406654343808,
|
| 564 |
+
"loss": 0.0107,
|
| 565 |
"step": 1900
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 3.53,
|
| 569 |
+
"grad_norm": 0.015557551756501198,
|
| 570 |
+
"learning_rate": 0.0006534195933456563,
|
| 571 |
+
"loss": 0.0104,
|
| 572 |
"step": 1925
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 3.57,
|
| 576 |
+
"grad_norm": 0.021405475214123726,
|
| 577 |
+
"learning_rate": 0.0006487985212569316,
|
| 578 |
+
"loss": 0.0117,
|
| 579 |
"step": 1950
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 3.62,
|
| 583 |
+
"grad_norm": 0.0015239958884194493,
|
| 584 |
+
"learning_rate": 0.0006441774491682071,
|
| 585 |
+
"loss": 0.0176,
|
| 586 |
"step": 1975
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 3.66,
|
| 590 |
+
"grad_norm": 0.0997876301407814,
|
| 591 |
+
"learning_rate": 0.0006395563770794824,
|
| 592 |
+
"loss": 0.0183,
|
| 593 |
"step": 2000
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 3.71,
|
| 597 |
+
"grad_norm": 0.004715020768344402,
|
| 598 |
+
"learning_rate": 0.0006349353049907579,
|
| 599 |
+
"loss": 0.0199,
|
| 600 |
"step": 2025
|
| 601 |
},
|
| 602 |
{
|
| 603 |
"epoch": 3.75,
|
| 604 |
+
"grad_norm": 0.1075858548283577,
|
| 605 |
+
"learning_rate": 0.0006303142329020333,
|
| 606 |
+
"loss": 0.0201,
|
| 607 |
"step": 2050
|
| 608 |
},
|
| 609 |
{
|
| 610 |
"epoch": 3.8,
|
| 611 |
+
"grad_norm": 0.020496558398008347,
|
| 612 |
+
"learning_rate": 0.0006256931608133087,
|
| 613 |
+
"loss": 0.0145,
|
| 614 |
"step": 2075
|
| 615 |
},
|
| 616 |
{
|
| 617 |
"epoch": 3.85,
|
| 618 |
+
"grad_norm": 0.11063025891780853,
|
| 619 |
+
"learning_rate": 0.0006210720887245841,
|
| 620 |
"loss": 0.0201,
|
| 621 |
"step": 2100
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 3.89,
|
| 625 |
+
"grad_norm": 0.1012192815542221,
|
| 626 |
+
"learning_rate": 0.0006164510166358595,
|
| 627 |
+
"loss": 0.0223,
|
| 628 |
"step": 2125
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 3.94,
|
| 632 |
+
"grad_norm": 0.04694315418601036,
|
| 633 |
+
"learning_rate": 0.0006118299445471349,
|
| 634 |
+
"loss": 0.0163,
|
| 635 |
"step": 2150
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 3.98,
|
| 639 |
+
"grad_norm": 0.05395512282848358,
|
| 640 |
+
"learning_rate": 0.0006072088724584103,
|
| 641 |
+
"loss": 0.009,
|
| 642 |
"step": 2175
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 4.0,
|
| 646 |
+
"eval_loss": 0.20004291832447052,
|
| 647 |
+
"eval_runtime": 464.3668,
|
| 648 |
+
"eval_samples_per_second": 1.751,
|
| 649 |
+
"eval_steps_per_second": 0.293,
|
| 650 |
"step": 2184
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 4.03,
|
| 654 |
+
"grad_norm": 0.08517912030220032,
|
| 655 |
+
"learning_rate": 0.0006025878003696857,
|
| 656 |
+
"loss": 0.0152,
|
| 657 |
"step": 2200
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 4.08,
|
| 661 |
+
"grad_norm": 0.23693686723709106,
|
| 662 |
+
"learning_rate": 0.0005979667282809613,
|
| 663 |
+
"loss": 0.0123,
|
| 664 |
"step": 2225
|
| 665 |
},
|
| 666 |
{
|
| 667 |
"epoch": 4.12,
|
| 668 |
+
"grad_norm": 0.04390133172273636,
|
| 669 |
+
"learning_rate": 0.0005933456561922366,
|
| 670 |
+
"loss": 0.0104,
|
| 671 |
"step": 2250
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 4.17,
|
| 675 |
+
"grad_norm": 0.048480305820703506,
|
| 676 |
+
"learning_rate": 0.0005887245841035121,
|
| 677 |
+
"loss": 0.0191,
|
| 678 |
"step": 2275
|
| 679 |
},
|
| 680 |
{
|
| 681 |
"epoch": 4.21,
|
| 682 |
+
"grad_norm": 0.07334431260824203,
|
| 683 |
+
"learning_rate": 0.0005841035120147874,
|
| 684 |
+
"loss": 0.0079,
|
| 685 |
"step": 2300
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 4.26,
|
| 689 |
+
"grad_norm": 0.26686009764671326,
|
| 690 |
+
"learning_rate": 0.0005794824399260629,
|
| 691 |
+
"loss": 0.0134,
|
| 692 |
"step": 2325
|
| 693 |
},
|
| 694 |
{
|
| 695 |
"epoch": 4.3,
|
| 696 |
+
"grad_norm": 0.18834412097930908,
|
| 697 |
+
"learning_rate": 0.0005748613678373382,
|
| 698 |
+
"loss": 0.0108,
|
| 699 |
"step": 2350
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 4.35,
|
| 703 |
+
"grad_norm": 0.11365604400634766,
|
| 704 |
+
"learning_rate": 0.0005702402957486137,
|
| 705 |
+
"loss": 0.0116,
|
| 706 |
"step": 2375
|
| 707 |
},
|
| 708 |
{
|
| 709 |
"epoch": 4.4,
|
| 710 |
+
"grad_norm": 0.21077445149421692,
|
| 711 |
+
"learning_rate": 0.0005656192236598892,
|
| 712 |
+
"loss": 0.017,
|
| 713 |
"step": 2400
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 4.44,
|
| 717 |
+
"grad_norm": 0.14450936019420624,
|
| 718 |
+
"learning_rate": 0.0005609981515711645,
|
| 719 |
+
"loss": 0.0056,
|
| 720 |
"step": 2425
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 4.49,
|
| 724 |
+
"grad_norm": 0.07659462839365005,
|
| 725 |
+
"learning_rate": 0.00055637707948244,
|
| 726 |
+
"loss": 0.0128,
|
| 727 |
"step": 2450
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 4.53,
|
| 731 |
+
"grad_norm": 0.07819797843694687,
|
| 732 |
+
"learning_rate": 0.0005517560073937153,
|
| 733 |
+
"loss": 0.0085,
|
| 734 |
"step": 2475
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 4.58,
|
| 738 |
+
"grad_norm": 0.10529200732707977,
|
| 739 |
+
"learning_rate": 0.0005471349353049908,
|
| 740 |
"loss": 0.0156,
|
| 741 |
"step": 2500
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 4.62,
|
| 745 |
+
"grad_norm": 0.034541305154561996,
|
| 746 |
+
"learning_rate": 0.0005425138632162662,
|
| 747 |
+
"loss": 0.0114,
|
| 748 |
"step": 2525
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 4.67,
|
| 752 |
+
"grad_norm": 0.0043388293124735355,
|
| 753 |
+
"learning_rate": 0.0005378927911275416,
|
| 754 |
+
"loss": 0.0114,
|
| 755 |
"step": 2550
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 4.72,
|
| 759 |
+
"grad_norm": 0.09843795001506805,
|
| 760 |
+
"learning_rate": 0.000533271719038817,
|
| 761 |
+
"loss": 0.0097,
|
| 762 |
"step": 2575
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 4.76,
|
| 766 |
+
"grad_norm": 0.1924191564321518,
|
| 767 |
+
"learning_rate": 0.0005286506469500924,
|
| 768 |
+
"loss": 0.0138,
|
| 769 |
"step": 2600
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 4.81,
|
| 773 |
+
"grad_norm": 0.0032940045930445194,
|
| 774 |
+
"learning_rate": 0.0005240295748613678,
|
| 775 |
+
"loss": 0.009,
|
| 776 |
"step": 2625
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 4.85,
|
| 780 |
+
"grad_norm": 0.17411276698112488,
|
| 781 |
+
"learning_rate": 0.0005194085027726432,
|
| 782 |
+
"loss": 0.005,
|
| 783 |
"step": 2650
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 4.9,
|
| 787 |
+
"grad_norm": 0.0008068850729614496,
|
| 788 |
+
"learning_rate": 0.0005147874306839186,
|
| 789 |
+
"loss": 0.0091,
|
| 790 |
"step": 2675
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 4.95,
|
| 794 |
+
"grad_norm": 0.013785873539745808,
|
| 795 |
+
"learning_rate": 0.0005101663585951941,
|
| 796 |
+
"loss": 0.0174,
|
| 797 |
"step": 2700
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 4.99,
|
| 801 |
+
"grad_norm": 0.06957102566957474,
|
| 802 |
+
"learning_rate": 0.0005055452865064695,
|
| 803 |
+
"loss": 0.0196,
|
| 804 |
"step": 2725
|
| 805 |
},
|
| 806 |
{
|
| 807 |
"epoch": 5.0,
|
| 808 |
+
"eval_loss": 0.1927657425403595,
|
| 809 |
+
"eval_runtime": 464.5332,
|
| 810 |
+
"eval_samples_per_second": 1.75,
|
| 811 |
+
"eval_steps_per_second": 0.293,
|
| 812 |
"step": 2730
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 5.04,
|
| 816 |
+
"grad_norm": 0.1873362511396408,
|
| 817 |
+
"learning_rate": 0.000500924214417745,
|
| 818 |
+
"loss": 0.0114,
|
| 819 |
"step": 2750
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 5.08,
|
| 823 |
+
"grad_norm": 0.013944294303655624,
|
| 824 |
+
"learning_rate": 0.0004963031423290203,
|
| 825 |
+
"loss": 0.0047,
|
| 826 |
"step": 2775
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 5.13,
|
| 830 |
+
"grad_norm": 0.14739681780338287,
|
| 831 |
+
"learning_rate": 0.0004916820702402958,
|
| 832 |
+
"loss": 0.0064,
|
| 833 |
"step": 2800
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 5.17,
|
| 837 |
+
"grad_norm": 0.039295587688684464,
|
| 838 |
+
"learning_rate": 0.00048706099815157115,
|
| 839 |
+
"loss": 0.0061,
|
| 840 |
"step": 2825
|
| 841 |
},
|
| 842 |
{
|
| 843 |
"epoch": 5.22,
|
| 844 |
+
"grad_norm": 0.009731476195156574,
|
| 845 |
+
"learning_rate": 0.0004824399260628466,
|
| 846 |
+
"loss": 0.0064,
|
| 847 |
"step": 2850
|
| 848 |
},
|
| 849 |
{
|
| 850 |
"epoch": 5.27,
|
| 851 |
+
"grad_norm": 0.009130421094596386,
|
| 852 |
+
"learning_rate": 0.000477818853974122,
|
| 853 |
+
"loss": 0.0056,
|
| 854 |
"step": 2875
|
| 855 |
},
|
| 856 |
{
|
| 857 |
"epoch": 5.31,
|
| 858 |
+
"grad_norm": 0.10517439246177673,
|
| 859 |
+
"learning_rate": 0.0004731977818853974,
|
| 860 |
+
"loss": 0.0095,
|
| 861 |
"step": 2900
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 5.36,
|
| 865 |
+
"grad_norm": 0.03147244080901146,
|
| 866 |
+
"learning_rate": 0.00046857670979667283,
|
| 867 |
+
"loss": 0.0069,
|
| 868 |
"step": 2925
|
| 869 |
},
|
| 870 |
{
|
| 871 |
"epoch": 5.4,
|
| 872 |
+
"grad_norm": 0.07550155371427536,
|
| 873 |
+
"learning_rate": 0.00046395563770794824,
|
| 874 |
+
"loss": 0.0084,
|
| 875 |
"step": 2950
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 5.45,
|
| 879 |
+
"grad_norm": 0.09899873286485672,
|
| 880 |
+
"learning_rate": 0.00045933456561922365,
|
| 881 |
+
"loss": 0.0087,
|
| 882 |
"step": 2975
|
| 883 |
},
|
| 884 |
{
|
| 885 |
"epoch": 5.49,
|
| 886 |
+
"grad_norm": 0.062454238533973694,
|
| 887 |
+
"learning_rate": 0.00045471349353049906,
|
| 888 |
+
"loss": 0.0114,
|
| 889 |
"step": 3000
|
| 890 |
},
|
| 891 |
{
|
| 892 |
"epoch": 5.54,
|
| 893 |
+
"grad_norm": 0.14996998012065887,
|
| 894 |
+
"learning_rate": 0.00045009242144177446,
|
| 895 |
+
"loss": 0.0091,
|
| 896 |
"step": 3025
|
| 897 |
},
|
| 898 |
{
|
| 899 |
"epoch": 5.59,
|
| 900 |
+
"grad_norm": 0.19108814001083374,
|
| 901 |
+
"learning_rate": 0.00044547134935304987,
|
| 902 |
+
"loss": 0.0147,
|
| 903 |
"step": 3050
|
| 904 |
},
|
| 905 |
{
|
| 906 |
"epoch": 5.63,
|
| 907 |
+
"grad_norm": 0.14450325071811676,
|
| 908 |
+
"learning_rate": 0.00044085027726432533,
|
| 909 |
+
"loss": 0.0152,
|
| 910 |
"step": 3075
|
| 911 |
},
|
| 912 |
{
|
| 913 |
"epoch": 5.68,
|
| 914 |
+
"grad_norm": 0.04423892870545387,
|
| 915 |
+
"learning_rate": 0.0004362292051756008,
|
| 916 |
+
"loss": 0.006,
|
| 917 |
"step": 3100
|
| 918 |
},
|
| 919 |
{
|
| 920 |
"epoch": 5.72,
|
| 921 |
+
"grad_norm": 0.13844439387321472,
|
| 922 |
+
"learning_rate": 0.0004316081330868762,
|
| 923 |
+
"loss": 0.009,
|
| 924 |
"step": 3125
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 5.77,
|
| 928 |
+
"grad_norm": 0.0006735218339599669,
|
| 929 |
+
"learning_rate": 0.0004269870609981516,
|
| 930 |
+
"loss": 0.0058,
|
| 931 |
"step": 3150
|
| 932 |
},
|
| 933 |
{
|
| 934 |
"epoch": 5.82,
|
| 935 |
+
"grad_norm": 0.011760660447180271,
|
| 936 |
+
"learning_rate": 0.000422365988909427,
|
| 937 |
+
"loss": 0.0049,
|
| 938 |
"step": 3175
|
| 939 |
},
|
| 940 |
{
|
| 941 |
"epoch": 5.86,
|
| 942 |
+
"grad_norm": 0.08969856053590775,
|
| 943 |
+
"learning_rate": 0.0004177449168207024,
|
| 944 |
+
"loss": 0.0065,
|
| 945 |
"step": 3200
|
| 946 |
},
|
| 947 |
{
|
| 948 |
"epoch": 5.91,
|
| 949 |
+
"grad_norm": 0.12556907534599304,
|
| 950 |
+
"learning_rate": 0.00041312384473197783,
|
| 951 |
+
"loss": 0.0089,
|
| 952 |
"step": 3225
|
| 953 |
},
|
| 954 |
{
|
| 955 |
"epoch": 5.95,
|
| 956 |
+
"grad_norm": 0.017725255340337753,
|
| 957 |
+
"learning_rate": 0.00040850277264325324,
|
| 958 |
+
"loss": 0.0088,
|
| 959 |
"step": 3250
|
| 960 |
},
|
| 961 |
{
|
| 962 |
"epoch": 6.0,
|
| 963 |
+
"grad_norm": 0.009897828102111816,
|
| 964 |
+
"learning_rate": 0.00040388170055452864,
|
| 965 |
+
"loss": 0.0071,
|
| 966 |
"step": 3275
|
| 967 |
},
|
| 968 |
{
|
| 969 |
"epoch": 6.0,
|
| 970 |
+
"eval_loss": 0.20994354784488678,
|
| 971 |
+
"eval_runtime": 463.0078,
|
| 972 |
+
"eval_samples_per_second": 1.756,
|
| 973 |
+
"eval_steps_per_second": 0.294,
|
| 974 |
"step": 3276
|
| 975 |
},
|
| 976 |
{
|
| 977 |
"epoch": 6.04,
|
| 978 |
+
"grad_norm": 0.0028004159685224295,
|
| 979 |
+
"learning_rate": 0.00039926062846580405,
|
| 980 |
+
"loss": 0.0093,
|
| 981 |
"step": 3300
|
| 982 |
},
|
| 983 |
{
|
| 984 |
"epoch": 6.09,
|
| 985 |
+
"grad_norm": 0.10490375012159348,
|
| 986 |
+
"learning_rate": 0.0003946395563770795,
|
| 987 |
+
"loss": 0.0053,
|
| 988 |
"step": 3325
|
| 989 |
},
|
| 990 |
{
|
| 991 |
"epoch": 6.14,
|
| 992 |
+
"grad_norm": 0.019779745489358902,
|
| 993 |
+
"learning_rate": 0.0003900184842883549,
|
| 994 |
+
"loss": 0.0084,
|
| 995 |
"step": 3350
|
| 996 |
},
|
| 997 |
{
|
| 998 |
"epoch": 6.18,
|
| 999 |
+
"grad_norm": 0.00020589173072949052,
|
| 1000 |
+
"learning_rate": 0.00038539741219963033,
|
| 1001 |
+
"loss": 0.0029,
|
| 1002 |
"step": 3375
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
"epoch": 6.23,
|
| 1006 |
+
"grad_norm": 0.003221085062250495,
|
| 1007 |
+
"learning_rate": 0.00038077634011090574,
|
| 1008 |
+
"loss": 0.0051,
|
| 1009 |
"step": 3400
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
"epoch": 6.27,
|
| 1013 |
+
"grad_norm": 0.00455264188349247,
|
| 1014 |
+
"learning_rate": 0.00037615526802218114,
|
| 1015 |
+
"loss": 0.0063,
|
| 1016 |
"step": 3425
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
"epoch": 6.32,
|
| 1020 |
+
"grad_norm": 0.00967650581151247,
|
| 1021 |
+
"learning_rate": 0.00037153419593345655,
|
| 1022 |
+
"loss": 0.0035,
|
| 1023 |
"step": 3450
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
"epoch": 6.36,
|
| 1027 |
+
"grad_norm": 0.009352604858577251,
|
| 1028 |
+
"learning_rate": 0.00036691312384473196,
|
| 1029 |
+
"loss": 0.0065,
|
| 1030 |
"step": 3475
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
"epoch": 6.41,
|
| 1034 |
+
"grad_norm": 0.002876508515328169,
|
| 1035 |
+
"learning_rate": 0.00036229205175600736,
|
| 1036 |
+
"loss": 0.0049,
|
| 1037 |
"step": 3500
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
"epoch": 6.46,
|
| 1041 |
+
"grad_norm": 0.002914861775934696,
|
| 1042 |
+
"learning_rate": 0.00035767097966728277,
|
| 1043 |
+
"loss": 0.0043,
|
| 1044 |
"step": 3525
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
"epoch": 6.5,
|
| 1048 |
+
"grad_norm": 0.021481545642018318,
|
| 1049 |
+
"learning_rate": 0.0003530499075785583,
|
| 1050 |
+
"loss": 0.0072,
|
| 1051 |
"step": 3550
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
"epoch": 6.55,
|
| 1055 |
+
"grad_norm": 0.08110266923904419,
|
| 1056 |
+
"learning_rate": 0.0003484288354898337,
|
| 1057 |
+
"loss": 0.0044,
|
| 1058 |
"step": 3575
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
"epoch": 6.59,
|
| 1062 |
+
"grad_norm": 0.020943278446793556,
|
| 1063 |
+
"learning_rate": 0.0003438077634011091,
|
| 1064 |
+
"loss": 0.0072,
|
| 1065 |
"step": 3600
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
"epoch": 6.64,
|
| 1069 |
+
"grad_norm": 0.005692564882338047,
|
| 1070 |
+
"learning_rate": 0.0003391866913123845,
|
| 1071 |
+
"loss": 0.0078,
|
| 1072 |
"step": 3625
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
"epoch": 6.68,
|
| 1076 |
+
"grad_norm": 0.11609622091054916,
|
| 1077 |
+
"learning_rate": 0.0003345656192236599,
|
| 1078 |
+
"loss": 0.008,
|
| 1079 |
"step": 3650
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
"epoch": 6.73,
|
| 1083 |
+
"grad_norm": 0.05904560536146164,
|
| 1084 |
+
"learning_rate": 0.0003299445471349353,
|
| 1085 |
+
"loss": 0.0061,
|
| 1086 |
"step": 3675
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
"epoch": 6.78,
|
| 1090 |
+
"grad_norm": 0.03346557542681694,
|
| 1091 |
+
"learning_rate": 0.00032532347504621073,
|
| 1092 |
+
"loss": 0.0069,
|
| 1093 |
"step": 3700
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
"epoch": 6.82,
|
| 1097 |
+
"grad_norm": 0.04848520830273628,
|
| 1098 |
+
"learning_rate": 0.00032070240295748614,
|
| 1099 |
+
"loss": 0.0078,
|
| 1100 |
"step": 3725
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
"epoch": 6.87,
|
| 1104 |
+
"grad_norm": 0.11064545810222626,
|
| 1105 |
+
"learning_rate": 0.00031608133086876155,
|
| 1106 |
+
"loss": 0.0083,
|
| 1107 |
"step": 3750
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
"epoch": 6.91,
|
| 1111 |
+
"grad_norm": 0.001821186626330018,
|
| 1112 |
+
"learning_rate": 0.00031146025878003695,
|
| 1113 |
+
"loss": 0.0084,
|
| 1114 |
"step": 3775
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
"epoch": 6.96,
|
| 1118 |
+
"grad_norm": 0.03919747844338417,
|
| 1119 |
+
"learning_rate": 0.0003068391866913124,
|
| 1120 |
+
"loss": 0.0054,
|
| 1121 |
"step": 3800
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
"epoch": 7.0,
|
| 1125 |
+
"eval_loss": 0.20703129470348358,
|
| 1126 |
+
"eval_runtime": 463.5868,
|
| 1127 |
+
"eval_samples_per_second": 1.754,
|
| 1128 |
+
"eval_steps_per_second": 0.293,
|
| 1129 |
"step": 3822
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 7.01,
|
| 1133 |
+
"grad_norm": 0.008144889958202839,
|
| 1134 |
+
"learning_rate": 0.0003022181146025878,
|
| 1135 |
+
"loss": 0.0057,
|
| 1136 |
"step": 3825
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
"epoch": 7.05,
|
| 1140 |
+
"grad_norm": 0.005378293804824352,
|
| 1141 |
+
"learning_rate": 0.00029759704251386323,
|
| 1142 |
+
"loss": 0.004,
|
| 1143 |
"step": 3850
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
"epoch": 7.1,
|
| 1147 |
+
"grad_norm": 0.03501349315047264,
|
| 1148 |
+
"learning_rate": 0.00029297597042513864,
|
| 1149 |
+
"loss": 0.003,
|
| 1150 |
"step": 3875
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
"epoch": 7.14,
|
| 1154 |
+
"grad_norm": 0.07073014974594116,
|
| 1155 |
+
"learning_rate": 0.00028835489833641404,
|
| 1156 |
+
"loss": 0.0029,
|
| 1157 |
"step": 3900
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 7.19,
|
| 1161 |
+
"grad_norm": 0.09017951786518097,
|
| 1162 |
+
"learning_rate": 0.00028373382624768945,
|
| 1163 |
+
"loss": 0.0027,
|
| 1164 |
"step": 3925
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
"epoch": 7.23,
|
| 1168 |
+
"grad_norm": 0.009881277568638325,
|
| 1169 |
+
"learning_rate": 0.00027911275415896486,
|
| 1170 |
+
"loss": 0.0044,
|
| 1171 |
"step": 3950
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 7.28,
|
| 1175 |
+
"grad_norm": 0.0018990118987858295,
|
| 1176 |
+
"learning_rate": 0.00027449168207024027,
|
| 1177 |
+
"loss": 0.0031,
|
| 1178 |
"step": 3975
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 7.33,
|
| 1182 |
+
"grad_norm": 0.004116680007427931,
|
| 1183 |
+
"learning_rate": 0.00026987060998151567,
|
| 1184 |
+
"loss": 0.0026,
|
| 1185 |
"step": 4000
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 7.37,
|
| 1189 |
+
"grad_norm": 0.03917807340621948,
|
| 1190 |
+
"learning_rate": 0.00026524953789279113,
|
| 1191 |
+
"loss": 0.0038,
|
| 1192 |
"step": 4025
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
"epoch": 7.42,
|
| 1196 |
+
"grad_norm": 0.0030583201441913843,
|
| 1197 |
+
"learning_rate": 0.0002606284658040666,
|
| 1198 |
+
"loss": 0.0032,
|
| 1199 |
"step": 4050
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 7.46,
|
| 1203 |
+
"grad_norm": 0.0014874553307890892,
|
| 1204 |
+
"learning_rate": 0.000256007393715342,
|
| 1205 |
+
"loss": 0.0054,
|
| 1206 |
"step": 4075
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
"epoch": 7.51,
|
| 1210 |
+
"grad_norm": 0.0008628646028228104,
|
| 1211 |
+
"learning_rate": 0.0002513863216266174,
|
| 1212 |
+
"loss": 0.0019,
|
| 1213 |
"step": 4100
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 7.55,
|
| 1217 |
+
"grad_norm": 0.02715575322508812,
|
| 1218 |
+
"learning_rate": 0.00024676524953789276,
|
| 1219 |
+
"loss": 0.0037,
|
| 1220 |
"step": 4125
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
"epoch": 7.6,
|
| 1224 |
+
"grad_norm": 0.0031906655058264732,
|
| 1225 |
+
"learning_rate": 0.00024214417744916822,
|
| 1226 |
+
"loss": 0.0058,
|
| 1227 |
"step": 4150
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 7.65,
|
| 1231 |
+
"grad_norm": 0.011863148771226406,
|
| 1232 |
+
"learning_rate": 0.00023752310536044363,
|
| 1233 |
+
"loss": 0.0022,
|
| 1234 |
"step": 4175
|
| 1235 |
},
|
| 1236 |
{
|
| 1237 |
"epoch": 7.69,
|
| 1238 |
+
"grad_norm": 0.0015202141366899014,
|
| 1239 |
+
"learning_rate": 0.00023290203327171904,
|
| 1240 |
+
"loss": 0.0045,
|
| 1241 |
"step": 4200
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 7.74,
|
| 1245 |
+
"grad_norm": 0.02240474335849285,
|
| 1246 |
+
"learning_rate": 0.00022828096118299447,
|
| 1247 |
+
"loss": 0.0039,
|
| 1248 |
"step": 4225
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
"epoch": 7.78,
|
| 1252 |
+
"grad_norm": 0.00918908603489399,
|
| 1253 |
+
"learning_rate": 0.00022365988909426988,
|
| 1254 |
+
"loss": 0.0014,
|
| 1255 |
"step": 4250
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
"epoch": 7.83,
|
| 1259 |
+
"grad_norm": 0.005950120277702808,
|
| 1260 |
+
"learning_rate": 0.0002190388170055453,
|
| 1261 |
+
"loss": 0.0031,
|
| 1262 |
"step": 4275
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
"epoch": 7.88,
|
| 1266 |
+
"grad_norm": 0.07433830946683884,
|
| 1267 |
+
"learning_rate": 0.0002144177449168207,
|
| 1268 |
+
"loss": 0.002,
|
| 1269 |
"step": 4300
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 7.92,
|
| 1273 |
+
"grad_norm": 0.09878811240196228,
|
| 1274 |
+
"learning_rate": 0.0002097966728280961,
|
| 1275 |
+
"loss": 0.0039,
|
| 1276 |
"step": 4325
|
| 1277 |
},
|
| 1278 |
{
|
| 1279 |
"epoch": 7.97,
|
| 1280 |
+
"grad_norm": 0.004627088084816933,
|
| 1281 |
+
"learning_rate": 0.00020517560073937154,
|
| 1282 |
+
"loss": 0.0066,
|
| 1283 |
"step": 4350
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
"epoch": 8.0,
|
| 1287 |
+
"eval_loss": 0.21887589991092682,
|
| 1288 |
+
"eval_runtime": 464.2458,
|
| 1289 |
+
"eval_samples_per_second": 1.751,
|
| 1290 |
+
"eval_steps_per_second": 0.293,
|
| 1291 |
"step": 4368
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
"epoch": 8.01,
|
| 1295 |
+
"grad_norm": 0.016953645274043083,
|
| 1296 |
+
"learning_rate": 0.00020055452865064697,
|
| 1297 |
+
"loss": 0.002,
|
| 1298 |
"step": 4375
|
| 1299 |
},
|
| 1300 |
{
|
| 1301 |
"epoch": 8.06,
|
| 1302 |
+
"grad_norm": 0.00016232863708864897,
|
| 1303 |
+
"learning_rate": 0.00019593345656192238,
|
| 1304 |
+
"loss": 0.0022,
|
| 1305 |
"step": 4400
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
"epoch": 8.1,
|
| 1309 |
+
"grad_norm": 0.00045125139877200127,
|
| 1310 |
+
"learning_rate": 0.00019131238447319779,
|
| 1311 |
+
"loss": 0.0016,
|
| 1312 |
"step": 4425
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
"epoch": 8.15,
|
| 1316 |
+
"grad_norm": 0.02065761759877205,
|
| 1317 |
+
"learning_rate": 0.0001866913123844732,
|
| 1318 |
+
"loss": 0.0017,
|
| 1319 |
"step": 4450
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
"epoch": 8.2,
|
| 1323 |
+
"grad_norm": 0.042185261845588684,
|
| 1324 |
+
"learning_rate": 0.00018207024029574863,
|
| 1325 |
+
"loss": 0.0027,
|
| 1326 |
"step": 4475
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
"epoch": 8.24,
|
| 1330 |
+
"grad_norm": 0.003087196499109268,
|
| 1331 |
+
"learning_rate": 0.00017744916820702404,
|
| 1332 |
+
"loss": 0.0018,
|
| 1333 |
"step": 4500
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
"epoch": 8.29,
|
| 1337 |
+
"grad_norm": 0.02859407104551792,
|
| 1338 |
+
"learning_rate": 0.00017282809611829944,
|
| 1339 |
+
"loss": 0.0015,
|
| 1340 |
"step": 4525
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
"epoch": 8.33,
|
| 1344 |
+
"grad_norm": 0.00041793755372054875,
|
| 1345 |
+
"learning_rate": 0.00016820702402957485,
|
| 1346 |
+
"loss": 0.0035,
|
| 1347 |
"step": 4550
|
| 1348 |
},
|
| 1349 |
{
|
| 1350 |
"epoch": 8.38,
|
| 1351 |
+
"grad_norm": 0.0037734461948275566,
|
| 1352 |
+
"learning_rate": 0.00016358595194085026,
|
| 1353 |
+
"loss": 0.002,
|
| 1354 |
"step": 4575
|
| 1355 |
},
|
| 1356 |
{
|
| 1357 |
"epoch": 8.42,
|
| 1358 |
+
"grad_norm": 0.0030207443051040173,
|
| 1359 |
+
"learning_rate": 0.00015896487985212572,
|
| 1360 |
+
"loss": 0.0022,
|
| 1361 |
"step": 4600
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
"epoch": 8.47,
|
| 1365 |
+
"grad_norm": 0.0026946039870381355,
|
| 1366 |
+
"learning_rate": 0.00015434380776340113,
|
| 1367 |
+
"loss": 0.0028,
|
| 1368 |
"step": 4625
|
| 1369 |
},
|
| 1370 |
{
|
| 1371 |
"epoch": 8.52,
|
| 1372 |
+
"grad_norm": 0.041892848908901215,
|
| 1373 |
+
"learning_rate": 0.00014972273567467653,
|
| 1374 |
+
"loss": 0.001,
|
| 1375 |
"step": 4650
|
| 1376 |
},
|
| 1377 |
{
|
| 1378 |
"epoch": 8.56,
|
| 1379 |
+
"grad_norm": 0.06906843930482864,
|
| 1380 |
+
"learning_rate": 0.00014510166358595194,
|
| 1381 |
+
"loss": 0.0013,
|
| 1382 |
"step": 4675
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
"epoch": 8.61,
|
| 1386 |
+
"grad_norm": 0.00067297019995749,
|
| 1387 |
+
"learning_rate": 0.00014048059149722737,
|
| 1388 |
+
"loss": 0.0029,
|
| 1389 |
"step": 4700
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
"epoch": 8.65,
|
| 1393 |
+
"grad_norm": 0.011746911332011223,
|
| 1394 |
+
"learning_rate": 0.00013585951940850278,
|
| 1395 |
+
"loss": 0.0012,
|
| 1396 |
"step": 4725
|
| 1397 |
},
|
| 1398 |
{
|
| 1399 |
"epoch": 8.7,
|
| 1400 |
+
"grad_norm": 0.0013995037879794836,
|
| 1401 |
+
"learning_rate": 0.0001312384473197782,
|
| 1402 |
+
"loss": 0.0017,
|
| 1403 |
"step": 4750
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
"epoch": 8.75,
|
| 1407 |
+
"grad_norm": 0.009580204263329506,
|
| 1408 |
+
"learning_rate": 0.0001266173752310536,
|
| 1409 |
+
"loss": 0.001,
|
| 1410 |
"step": 4775
|
| 1411 |
},
|
| 1412 |
{
|
| 1413 |
"epoch": 8.79,
|
| 1414 |
+
"grad_norm": 0.0008843488758429885,
|
| 1415 |
+
"learning_rate": 0.00012199630314232903,
|
| 1416 |
+
"loss": 0.0015,
|
| 1417 |
"step": 4800
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
"epoch": 8.84,
|
| 1421 |
+
"grad_norm": 0.0013571062590926886,
|
| 1422 |
+
"learning_rate": 0.00011737523105360444,
|
| 1423 |
+
"loss": 0.0024,
|
| 1424 |
"step": 4825
|
| 1425 |
},
|
| 1426 |
{
|
| 1427 |
"epoch": 8.88,
|
| 1428 |
+
"grad_norm": 0.01475840900093317,
|
| 1429 |
+
"learning_rate": 0.00011275415896487985,
|
| 1430 |
+
"loss": 0.0009,
|
| 1431 |
"step": 4850
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
"epoch": 8.93,
|
| 1435 |
+
"grad_norm": 0.008486463688313961,
|
| 1436 |
+
"learning_rate": 0.00010813308687615527,
|
| 1437 |
+
"loss": 0.0022,
|
| 1438 |
"step": 4875
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
"epoch": 8.97,
|
| 1442 |
+
"grad_norm": 0.0005981879075989127,
|
| 1443 |
+
"learning_rate": 0.00010351201478743069,
|
| 1444 |
+
"loss": 0.0006,
|
| 1445 |
"step": 4900
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
"epoch": 9.0,
|
| 1449 |
+
"eval_loss": 0.23254649341106415,
|
| 1450 |
+
"eval_runtime": 463.7576,
|
| 1451 |
+
"eval_samples_per_second": 1.753,
|
| 1452 |
+
"eval_steps_per_second": 0.293,
|
| 1453 |
"step": 4914
|
| 1454 |
},
|
| 1455 |
{
|
| 1456 |
"epoch": 9.02,
|
| 1457 |
+
"grad_norm": 0.011784604750573635,
|
| 1458 |
+
"learning_rate": 9.889094269870611e-05,
|
| 1459 |
+
"loss": 0.0028,
|
| 1460 |
"step": 4925
|
| 1461 |
},
|
| 1462 |
{
|
| 1463 |
"epoch": 9.07,
|
| 1464 |
+
"grad_norm": 0.0005488657625392079,
|
| 1465 |
+
"learning_rate": 9.426987060998152e-05,
|
| 1466 |
+
"loss": 0.0016,
|
| 1467 |
"step": 4950
|
| 1468 |
},
|
| 1469 |
{
|
| 1470 |
"epoch": 9.11,
|
| 1471 |
+
"grad_norm": 0.0024228901602327824,
|
| 1472 |
+
"learning_rate": 8.964879852125694e-05,
|
| 1473 |
+
"loss": 0.001,
|
| 1474 |
"step": 4975
|
| 1475 |
},
|
| 1476 |
{
|
| 1477 |
"epoch": 9.16,
|
| 1478 |
+
"grad_norm": 0.0021140037570148706,
|
| 1479 |
+
"learning_rate": 8.502772643253234e-05,
|
| 1480 |
+
"loss": 0.0014,
|
| 1481 |
"step": 5000
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
"epoch": 9.2,
|
| 1485 |
+
"grad_norm": 0.0011844311375170946,
|
| 1486 |
+
"learning_rate": 8.040665434380776e-05,
|
| 1487 |
+
"loss": 0.001,
|
| 1488 |
"step": 5025
|
| 1489 |
},
|
| 1490 |
{
|
| 1491 |
"epoch": 9.25,
|
| 1492 |
+
"grad_norm": 0.011841055937111378,
|
| 1493 |
+
"learning_rate": 7.578558225508319e-05,
|
| 1494 |
+
"loss": 0.0009,
|
| 1495 |
"step": 5050
|
| 1496 |
},
|
| 1497 |
{
|
| 1498 |
"epoch": 9.29,
|
| 1499 |
+
"grad_norm": 0.013395372778177261,
|
| 1500 |
+
"learning_rate": 7.116451016635859e-05,
|
| 1501 |
+
"loss": 0.0009,
|
| 1502 |
"step": 5075
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
"epoch": 9.34,
|
| 1506 |
+
"grad_norm": 0.05545121058821678,
|
| 1507 |
+
"learning_rate": 6.654343807763401e-05,
|
| 1508 |
+
"loss": 0.0012,
|
| 1509 |
"step": 5100
|
| 1510 |
},
|
| 1511 |
{
|
| 1512 |
"epoch": 9.39,
|
| 1513 |
+
"grad_norm": 0.01891588233411312,
|
| 1514 |
+
"learning_rate": 6.192236598890943e-05,
|
| 1515 |
+
"loss": 0.0006,
|
| 1516 |
"step": 5125
|
| 1517 |
},
|
| 1518 |
{
|
| 1519 |
"epoch": 9.43,
|
| 1520 |
+
"grad_norm": 0.0025335114914923906,
|
| 1521 |
+
"learning_rate": 5.730129390018484e-05,
|
| 1522 |
+
"loss": 0.0006,
|
| 1523 |
"step": 5150
|
| 1524 |
},
|
| 1525 |
{
|
| 1526 |
"epoch": 9.48,
|
| 1527 |
+
"grad_norm": 0.0021167423110455275,
|
| 1528 |
+
"learning_rate": 5.268022181146026e-05,
|
| 1529 |
+
"loss": 0.0007,
|
| 1530 |
"step": 5175
|
| 1531 |
},
|
| 1532 |
{
|
| 1533 |
"epoch": 9.52,
|
| 1534 |
+
"grad_norm": 0.0011415353510528803,
|
| 1535 |
+
"learning_rate": 4.8059149722735676e-05,
|
| 1536 |
+
"loss": 0.0014,
|
| 1537 |
"step": 5200
|
| 1538 |
},
|
| 1539 |
{
|
| 1540 |
"epoch": 9.57,
|
| 1541 |
+
"grad_norm": 0.00026013093884103,
|
| 1542 |
+
"learning_rate": 4.343807763401109e-05,
|
| 1543 |
+
"loss": 0.0007,
|
| 1544 |
"step": 5225
|
| 1545 |
},
|
| 1546 |
{
|
| 1547 |
"epoch": 9.62,
|
| 1548 |
+
"grad_norm": 0.03879648819565773,
|
| 1549 |
+
"learning_rate": 3.8817005545286504e-05,
|
| 1550 |
+
"loss": 0.0007,
|
| 1551 |
"step": 5250
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
"epoch": 9.66,
|
| 1555 |
+
"grad_norm": 0.006720875855535269,
|
| 1556 |
+
"learning_rate": 3.4195933456561925e-05,
|
| 1557 |
+
"loss": 0.0009,
|
| 1558 |
"step": 5275
|
| 1559 |
},
|
| 1560 |
{
|
| 1561 |
"epoch": 9.71,
|
| 1562 |
+
"grad_norm": 0.006371485069394112,
|
| 1563 |
+
"learning_rate": 2.957486136783734e-05,
|
| 1564 |
+
"loss": 0.0009,
|
| 1565 |
"step": 5300
|
| 1566 |
},
|
| 1567 |
{
|
| 1568 |
"epoch": 9.75,
|
| 1569 |
+
"grad_norm": 0.012291524559259415,
|
| 1570 |
+
"learning_rate": 2.4953789279112753e-05,
|
| 1571 |
+
"loss": 0.0012,
|
| 1572 |
"step": 5325
|
| 1573 |
},
|
| 1574 |
{
|
| 1575 |
"epoch": 9.8,
|
| 1576 |
+
"grad_norm": 0.012388636358082294,
|
| 1577 |
+
"learning_rate": 2.033271719038817e-05,
|
| 1578 |
+
"loss": 0.0006,
|
| 1579 |
"step": 5350
|
| 1580 |
},
|
| 1581 |
{
|
| 1582 |
"epoch": 9.84,
|
| 1583 |
+
"grad_norm": 0.0905984491109848,
|
| 1584 |
+
"learning_rate": 1.5711645101663588e-05,
|
| 1585 |
+
"loss": 0.0011,
|
| 1586 |
"step": 5375
|
| 1587 |
},
|
| 1588 |
{
|
| 1589 |
"epoch": 9.89,
|
| 1590 |
+
"grad_norm": 0.0024207117967307568,
|
| 1591 |
+
"learning_rate": 1.1090573012939002e-05,
|
| 1592 |
+
"loss": 0.001,
|
| 1593 |
"step": 5400
|
| 1594 |
},
|
| 1595 |
{
|
| 1596 |
"epoch": 9.94,
|
| 1597 |
+
"grad_norm": 0.003070174716413021,
|
| 1598 |
+
"learning_rate": 6.469500924214418e-06,
|
| 1599 |
+
"loss": 0.0008,
|
| 1600 |
"step": 5425
|
| 1601 |
},
|
| 1602 |
{
|
| 1603 |
"epoch": 9.98,
|
| 1604 |
+
"grad_norm": 0.012533812783658504,
|
| 1605 |
+
"learning_rate": 1.8484288354898337e-06,
|
| 1606 |
+
"loss": 0.001,
|
| 1607 |
"step": 5450
|
| 1608 |
},
|
| 1609 |
{
|
| 1610 |
"epoch": 10.0,
|
| 1611 |
+
"eval_loss": 0.23037216067314148,
|
| 1612 |
+
"eval_runtime": 463.1996,
|
| 1613 |
+
"eval_samples_per_second": 1.755,
|
| 1614 |
+
"eval_steps_per_second": 0.294,
|
| 1615 |
"step": 5460
|
| 1616 |
},
|
| 1617 |
{
|
| 1618 |
"epoch": 10.0,
|
| 1619 |
"step": 5460,
|
| 1620 |
"total_flos": 1.135723105419264e+20,
|
| 1621 |
+
"train_loss": 0.029495533068592733,
|
| 1622 |
+
"train_runtime": 29108.855,
|
| 1623 |
+
"train_samples_per_second": 1.124,
|
| 1624 |
"train_steps_per_second": 0.188
|
| 1625 |
}
|
| 1626 |
],
|