End of training
Browse files- all_results.json +10 -10
- eval_results.json +5 -5
- runs/Apr13_14-54-16_67071951de9d/events.out.tfevents.1713021034.67071951de9d.6427.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +612 -360
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.99,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos": 2.
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.99,
|
| 3 |
+
"eval_accuracy": 0.9844,
|
| 4 |
+
"eval_loss": 0.04590694606304169,
|
| 5 |
+
"eval_runtime": 32.7189,
|
| 6 |
+
"eval_samples_per_second": 305.634,
|
| 7 |
+
"eval_steps_per_second": 9.566,
|
| 8 |
+
"total_flos": 2.930358373492064e+18,
|
| 9 |
+
"train_loss": 0.33610059461023056,
|
| 10 |
+
"train_runtime": 935.657,
|
| 11 |
+
"train_samples_per_second": 160.315,
|
| 12 |
+
"train_steps_per_second": 1.25
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.99,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.99,
|
| 3 |
+
"eval_accuracy": 0.9844,
|
| 4 |
+
"eval_loss": 0.04590694606304169,
|
| 5 |
+
"eval_runtime": 32.7189,
|
| 6 |
+
"eval_samples_per_second": 305.634,
|
| 7 |
+
"eval_steps_per_second": 9.566
|
| 8 |
}
|
runs/Apr13_14-54-16_67071951de9d/events.out.tfevents.1713021034.67071951de9d.6427.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b142f82300c3b85bcbfdc744203b64d047b3b7d888be2ef12e6a2ec6e995de5
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.99,
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.99,
|
| 3 |
+
"total_flos": 2.930358373492064e+18,
|
| 4 |
+
"train_loss": 0.33610059461023056,
|
| 5 |
+
"train_runtime": 935.657,
|
| 6 |
+
"train_samples_per_second": 160.315,
|
| 7 |
+
"train_steps_per_second": 1.25
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,623 +1,875 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "vit-small-patch16-224-finetuned-cifar10/checkpoint-
|
| 4 |
-
"epoch": 2.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 7.
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 7.
|
| 28 |
-
"learning_rate": 1.
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 6.
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm": 6.
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm": 5.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm": 6.
|
| 77 |
-
"learning_rate": 4.
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm": 5.
|
| 84 |
-
"learning_rate": 4.
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm": 5.
|
| 91 |
-
"learning_rate": 4.
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate": 4.
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 4.
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm": 5.
|
| 112 |
-
"learning_rate": 4.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate": 4.
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate": 4.
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate": 4.
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate": 4.
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 190
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm": 5.
|
| 147 |
-
"learning_rate": 4.
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 200
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate": 4.
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 210
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm":
|
| 161 |
-
"learning_rate": 4.
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 220
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 230
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm":
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 240
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm":
|
| 182 |
-
"learning_rate":
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 250
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm":
|
| 189 |
-
"learning_rate":
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 260
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate":
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 270
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch":
|
| 202 |
-
"
|
| 203 |
-
"
|
| 204 |
-
"
|
| 205 |
-
"eval_samples_per_second": 336.613,
|
| 206 |
-
"eval_steps_per_second": 10.525,
|
| 207 |
-
"step": 273
|
| 208 |
-
},
|
| 209 |
-
{
|
| 210 |
-
"epoch": 1.02,
|
| 211 |
-
"grad_norm": 5.443094253540039,
|
| 212 |
-
"learning_rate": 3.656716417910448e-05,
|
| 213 |
-
"loss": 0.292,
|
| 214 |
"step": 280
|
| 215 |
},
|
| 216 |
{
|
| 217 |
-
"epoch":
|
| 218 |
-
"grad_norm":
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 290
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"grad_norm": 5.
|
| 226 |
-
"learning_rate":
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 300
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch":
|
| 232 |
-
"grad_norm": 4.
|
| 233 |
-
"learning_rate":
|
| 234 |
-
"loss": 0.
|
| 235 |
"step": 310
|
| 236 |
},
|
| 237 |
{
|
| 238 |
-
"epoch":
|
| 239 |
-
"grad_norm": 5.
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 320
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"grad_norm":
|
| 247 |
-
"learning_rate": 3.
|
| 248 |
-
"loss": 0.
|
| 249 |
"step": 330
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch":
|
| 253 |
-
"grad_norm": 4.
|
| 254 |
-
"learning_rate": 3.
|
| 255 |
-
"loss": 0.
|
| 256 |
"step": 340
|
| 257 |
},
|
| 258 |
{
|
| 259 |
-
"epoch":
|
| 260 |
-
"grad_norm":
|
| 261 |
-
"learning_rate": 3.
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 350
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"grad_norm":
|
| 268 |
-
"learning_rate": 3.
|
| 269 |
-
"loss": 0.
|
| 270 |
"step": 360
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch":
|
| 274 |
-
"grad_norm": 5.
|
| 275 |
-
"learning_rate": 3.
|
| 276 |
-
"loss": 0.
|
| 277 |
"step": 370
|
| 278 |
},
|
| 279 |
{
|
| 280 |
-
"epoch":
|
| 281 |
-
"grad_norm": 4.
|
| 282 |
-
"learning_rate":
|
| 283 |
-
"loss": 0.
|
| 284 |
"step": 380
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch": 1.
|
| 288 |
-
"grad_norm":
|
| 289 |
-
"learning_rate":
|
| 290 |
-
"loss": 0.
|
| 291 |
"step": 390
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch": 1.
|
| 295 |
-
"
|
| 296 |
-
"
|
| 297 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 298 |
"step": 400
|
| 299 |
},
|
| 300 |
{
|
| 301 |
-
"epoch": 1.
|
| 302 |
-
"grad_norm": 3.
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 410
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch": 1.
|
| 309 |
-
"grad_norm":
|
| 310 |
-
"learning_rate":
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 420
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch": 1.
|
| 316 |
-
"grad_norm":
|
| 317 |
-
"learning_rate":
|
| 318 |
-
"loss": 0.
|
| 319 |
"step": 430
|
| 320 |
},
|
| 321 |
{
|
| 322 |
-
"epoch": 1.
|
| 323 |
-
"grad_norm":
|
| 324 |
-
"learning_rate":
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 440
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch": 1.
|
| 330 |
-
"grad_norm": 5.
|
| 331 |
-
"learning_rate":
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 450
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch": 1.
|
| 337 |
-
"grad_norm":
|
| 338 |
-
"learning_rate":
|
| 339 |
-
"loss": 0.
|
| 340 |
"step": 460
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"epoch": 1.
|
| 344 |
-
"grad_norm":
|
| 345 |
-
"learning_rate":
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 470
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 1.
|
| 351 |
-
"grad_norm":
|
| 352 |
-
"learning_rate":
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 480
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 1.
|
| 358 |
-
"grad_norm":
|
| 359 |
-
"learning_rate":
|
| 360 |
-
"loss": 0.
|
| 361 |
"step": 490
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch": 1.
|
| 365 |
-
"grad_norm":
|
| 366 |
-
"learning_rate":
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 500
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch": 1.
|
| 372 |
-
"grad_norm": 4.
|
| 373 |
-
"learning_rate":
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 510
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch": 1.
|
| 379 |
-
"grad_norm":
|
| 380 |
-
"learning_rate":
|
| 381 |
-
"loss": 0.
|
| 382 |
"step": 520
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch": 1.
|
| 386 |
-
"grad_norm":
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 530
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 1.
|
| 393 |
-
"grad_norm":
|
| 394 |
-
"learning_rate":
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 540
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch":
|
| 400 |
-
"
|
| 401 |
-
"
|
| 402 |
-
"
|
| 403 |
-
"eval_samples_per_second": 333.446,
|
| 404 |
-
"eval_steps_per_second": 10.426,
|
| 405 |
-
"step": 547
|
| 406 |
-
},
|
| 407 |
-
{
|
| 408 |
-
"epoch": 2.01,
|
| 409 |
-
"grad_norm": 4.121983528137207,
|
| 410 |
-
"learning_rate": 1.824966078697422e-05,
|
| 411 |
-
"loss": 0.2469,
|
| 412 |
"step": 550
|
| 413 |
},
|
| 414 |
{
|
| 415 |
-
"epoch":
|
| 416 |
-
"grad_norm":
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss": 0.
|
| 419 |
"step": 560
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch":
|
| 423 |
-
"grad_norm":
|
| 424 |
-
"learning_rate":
|
| 425 |
-
"loss": 0.
|
| 426 |
"step": 570
|
| 427 |
},
|
| 428 |
{
|
| 429 |
-
"epoch":
|
| 430 |
-
"grad_norm":
|
| 431 |
-
"learning_rate":
|
| 432 |
-
"loss": 0.
|
| 433 |
"step": 580
|
| 434 |
},
|
| 435 |
{
|
| 436 |
-
"epoch":
|
| 437 |
-
"grad_norm": 4.
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss": 0.
|
| 440 |
"step": 590
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"epoch":
|
| 444 |
-
"grad_norm": 3.
|
| 445 |
-
"learning_rate":
|
| 446 |
-
"loss": 0.
|
| 447 |
"step": 600
|
| 448 |
},
|
| 449 |
{
|
| 450 |
-
"epoch":
|
| 451 |
-
"grad_norm":
|
| 452 |
-
"learning_rate":
|
| 453 |
-
"loss": 0.
|
| 454 |
"step": 610
|
| 455 |
},
|
| 456 |
{
|
| 457 |
-
"epoch":
|
| 458 |
-
"grad_norm":
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 620
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch":
|
| 465 |
-
"grad_norm":
|
| 466 |
-
"learning_rate":
|
| 467 |
-
"loss": 0.
|
| 468 |
"step": 630
|
| 469 |
},
|
| 470 |
{
|
| 471 |
-
"epoch":
|
| 472 |
-
"grad_norm":
|
| 473 |
-
"learning_rate":
|
| 474 |
-
"loss": 0.
|
| 475 |
"step": 640
|
| 476 |
},
|
| 477 |
{
|
| 478 |
-
"epoch":
|
| 479 |
-
"grad_norm":
|
| 480 |
-
"learning_rate":
|
| 481 |
-
"loss": 0.
|
| 482 |
"step": 650
|
| 483 |
},
|
| 484 |
{
|
| 485 |
-
"epoch":
|
| 486 |
-
"grad_norm":
|
| 487 |
-
"learning_rate":
|
| 488 |
-
"loss": 0.
|
| 489 |
"step": 660
|
| 490 |
},
|
| 491 |
{
|
| 492 |
-
"epoch":
|
| 493 |
-
"grad_norm":
|
| 494 |
-
"learning_rate":
|
| 495 |
-
"loss": 0.
|
| 496 |
"step": 670
|
| 497 |
},
|
| 498 |
{
|
| 499 |
-
"epoch":
|
| 500 |
-
"grad_norm": 4.
|
| 501 |
-
"learning_rate":
|
| 502 |
-
"loss": 0.
|
| 503 |
"step": 680
|
| 504 |
},
|
| 505 |
{
|
| 506 |
-
"epoch":
|
| 507 |
-
"grad_norm":
|
| 508 |
-
"learning_rate":
|
| 509 |
-
"loss": 0.
|
| 510 |
"step": 690
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch":
|
| 514 |
-
"grad_norm":
|
| 515 |
-
"learning_rate":
|
| 516 |
-
"loss": 0.
|
| 517 |
"step": 700
|
| 518 |
},
|
| 519 |
{
|
| 520 |
-
"epoch":
|
| 521 |
-
"grad_norm": 3.
|
| 522 |
-
"learning_rate":
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 710
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch":
|
| 528 |
-
"grad_norm":
|
| 529 |
-
"learning_rate":
|
| 530 |
-
"loss": 0.
|
| 531 |
"step": 720
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"epoch":
|
| 535 |
-
"grad_norm":
|
| 536 |
-
"learning_rate":
|
| 537 |
-
"loss": 0.
|
| 538 |
"step": 730
|
| 539 |
},
|
| 540 |
{
|
| 541 |
-
"epoch":
|
| 542 |
-
"grad_norm":
|
| 543 |
-
"learning_rate":
|
| 544 |
-
"loss": 0.
|
| 545 |
"step": 740
|
| 546 |
},
|
| 547 |
{
|
| 548 |
-
"epoch":
|
| 549 |
-
"grad_norm":
|
| 550 |
-
"learning_rate":
|
| 551 |
-
"loss": 0.
|
| 552 |
"step": 750
|
| 553 |
},
|
| 554 |
{
|
| 555 |
-
"epoch":
|
| 556 |
-
"grad_norm":
|
| 557 |
-
"learning_rate":
|
| 558 |
-
"loss": 0.
|
| 559 |
"step": 760
|
| 560 |
},
|
| 561 |
{
|
| 562 |
-
"epoch":
|
| 563 |
-
"grad_norm":
|
| 564 |
-
"learning_rate":
|
| 565 |
-
"loss": 0.
|
| 566 |
"step": 770
|
| 567 |
},
|
| 568 |
{
|
| 569 |
-
"epoch": 2.
|
| 570 |
-
"grad_norm":
|
| 571 |
-
"learning_rate":
|
| 572 |
-
"loss": 0.
|
| 573 |
"step": 780
|
| 574 |
},
|
| 575 |
{
|
| 576 |
-
"epoch": 2.
|
| 577 |
-
"
|
| 578 |
-
"
|
| 579 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 580 |
"step": 790
|
| 581 |
},
|
| 582 |
{
|
| 583 |
-
"epoch": 2.
|
| 584 |
-
"grad_norm": 4.
|
| 585 |
-
"learning_rate": 1.
|
| 586 |
-
"loss": 0.
|
| 587 |
"step": 800
|
| 588 |
},
|
| 589 |
{
|
| 590 |
-
"epoch": 2.
|
| 591 |
-
"grad_norm":
|
| 592 |
-
"learning_rate":
|
| 593 |
-
"loss": 0.
|
| 594 |
"step": 810
|
| 595 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 596 |
{
|
| 597 |
"epoch": 2.99,
|
| 598 |
-
"eval_accuracy": 0.
|
| 599 |
-
"eval_loss": 0.
|
| 600 |
-
"eval_runtime":
|
| 601 |
-
"eval_samples_per_second":
|
| 602 |
-
"eval_steps_per_second":
|
| 603 |
-
"step":
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 2.99,
|
| 607 |
-
"step":
|
| 608 |
-
"total_flos": 2.
|
| 609 |
-
"train_loss": 0.
|
| 610 |
-
"train_runtime":
|
| 611 |
-
"train_samples_per_second":
|
| 612 |
-
"train_steps_per_second": 1.
|
| 613 |
}
|
| 614 |
],
|
| 615 |
"logging_steps": 10,
|
| 616 |
-
"max_steps":
|
| 617 |
"num_input_tokens_seen": 0,
|
| 618 |
"num_train_epochs": 3,
|
| 619 |
"save_steps": 500,
|
| 620 |
-
"total_flos": 2.
|
| 621 |
"train_batch_size": 32,
|
| 622 |
"trial_name": null,
|
| 623 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9844,
|
| 3 |
+
"best_model_checkpoint": "vit-small-patch16-224-finetuned-cifar10/checkpoint-1170",
|
| 4 |
+
"epoch": 2.9942418426103647,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1170,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"grad_norm": 8.970051765441895,
|
| 14 |
+
"learning_rate": 4.273504273504274e-06,
|
| 15 |
+
"loss": 2.432,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.05,
|
| 20 |
+
"grad_norm": 7.947530746459961,
|
| 21 |
+
"learning_rate": 8.547008547008548e-06,
|
| 22 |
+
"loss": 2.2376,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.08,
|
| 27 |
+
"grad_norm": 7.60875129699707,
|
| 28 |
+
"learning_rate": 1.282051282051282e-05,
|
| 29 |
+
"loss": 1.9698,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.1,
|
| 34 |
+
"grad_norm": 6.434678554534912,
|
| 35 |
+
"learning_rate": 1.7094017094017095e-05,
|
| 36 |
+
"loss": 1.5489,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.13,
|
| 41 |
+
"grad_norm": 6.238937854766846,
|
| 42 |
+
"learning_rate": 2.1367521367521368e-05,
|
| 43 |
+
"loss": 1.1375,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.15,
|
| 48 |
+
"grad_norm": 5.423764228820801,
|
| 49 |
+
"learning_rate": 2.564102564102564e-05,
|
| 50 |
+
"loss": 0.7332,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.18,
|
| 55 |
+
"grad_norm": 7.145791530609131,
|
| 56 |
+
"learning_rate": 2.9914529914529915e-05,
|
| 57 |
+
"loss": 0.5841,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.2,
|
| 62 |
+
"grad_norm": 6.430414199829102,
|
| 63 |
+
"learning_rate": 3.418803418803419e-05,
|
| 64 |
+
"loss": 0.4957,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.23,
|
| 69 |
+
"grad_norm": 6.975243091583252,
|
| 70 |
+
"learning_rate": 3.846153846153846e-05,
|
| 71 |
+
"loss": 0.4523,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.26,
|
| 76 |
+
"grad_norm": 6.09722900390625,
|
| 77 |
+
"learning_rate": 4.2735042735042735e-05,
|
| 78 |
+
"loss": 0.4462,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.28,
|
| 83 |
+
"grad_norm": 5.052177429199219,
|
| 84 |
+
"learning_rate": 4.700854700854701e-05,
|
| 85 |
+
"loss": 0.3624,
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.31,
|
| 90 |
+
"grad_norm": 5.480886459350586,
|
| 91 |
+
"learning_rate": 4.985754985754986e-05,
|
| 92 |
+
"loss": 0.3895,
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.33,
|
| 97 |
+
"grad_norm": 4.691812992095947,
|
| 98 |
+
"learning_rate": 4.938271604938271e-05,
|
| 99 |
+
"loss": 0.3446,
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.36,
|
| 104 |
+
"grad_norm": 6.404294013977051,
|
| 105 |
+
"learning_rate": 4.890788224121557e-05,
|
| 106 |
+
"loss": 0.4006,
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.38,
|
| 111 |
+
"grad_norm": 5.33477258682251,
|
| 112 |
+
"learning_rate": 4.8433048433048433e-05,
|
| 113 |
+
"loss": 0.3532,
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.41,
|
| 118 |
+
"grad_norm": 5.476822853088379,
|
| 119 |
+
"learning_rate": 4.7958214624881294e-05,
|
| 120 |
+
"loss": 0.3421,
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.44,
|
| 125 |
+
"grad_norm": 6.005227565765381,
|
| 126 |
+
"learning_rate": 4.7483380816714154e-05,
|
| 127 |
+
"loss": 0.3692,
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.46,
|
| 132 |
+
"grad_norm": 4.84128999710083,
|
| 133 |
+
"learning_rate": 4.700854700854701e-05,
|
| 134 |
+
"loss": 0.3457,
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.49,
|
| 139 |
+
"grad_norm": 4.650127410888672,
|
| 140 |
+
"learning_rate": 4.653371320037987e-05,
|
| 141 |
+
"loss": 0.3685,
|
| 142 |
"step": 190
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.51,
|
| 146 |
+
"grad_norm": 5.779717445373535,
|
| 147 |
+
"learning_rate": 4.605887939221273e-05,
|
| 148 |
+
"loss": 0.3451,
|
| 149 |
"step": 200
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.54,
|
| 153 |
+
"grad_norm": 6.38178825378418,
|
| 154 |
+
"learning_rate": 4.558404558404559e-05,
|
| 155 |
+
"loss": 0.3394,
|
| 156 |
"step": 210
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.56,
|
| 160 |
+
"grad_norm": 5.148958206176758,
|
| 161 |
+
"learning_rate": 4.510921177587845e-05,
|
| 162 |
+
"loss": 0.3641,
|
| 163 |
"step": 220
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.59,
|
| 167 |
+
"grad_norm": 3.9889109134674072,
|
| 168 |
+
"learning_rate": 4.463437796771131e-05,
|
| 169 |
+
"loss": 0.3108,
|
| 170 |
"step": 230
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.61,
|
| 174 |
+
"grad_norm": 6.51384162902832,
|
| 175 |
+
"learning_rate": 4.415954415954416e-05,
|
| 176 |
+
"loss": 0.2949,
|
| 177 |
"step": 240
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.64,
|
| 181 |
+
"grad_norm": 4.950154781341553,
|
| 182 |
+
"learning_rate": 4.368471035137702e-05,
|
| 183 |
+
"loss": 0.3352,
|
| 184 |
"step": 250
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.67,
|
| 188 |
+
"grad_norm": 6.234986782073975,
|
| 189 |
+
"learning_rate": 4.3209876543209875e-05,
|
| 190 |
+
"loss": 0.309,
|
| 191 |
"step": 260
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.69,
|
| 195 |
+
"grad_norm": 5.57274055480957,
|
| 196 |
+
"learning_rate": 4.2735042735042735e-05,
|
| 197 |
+
"loss": 0.3161,
|
| 198 |
"step": 270
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.72,
|
| 202 |
+
"grad_norm": 5.511316776275635,
|
| 203 |
+
"learning_rate": 4.2260208926875595e-05,
|
| 204 |
+
"loss": 0.3018,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 205 |
"step": 280
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.74,
|
| 209 |
+
"grad_norm": 3.934835910797119,
|
| 210 |
+
"learning_rate": 4.1785375118708455e-05,
|
| 211 |
+
"loss": 0.3116,
|
| 212 |
"step": 290
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.77,
|
| 216 |
+
"grad_norm": 5.323722839355469,
|
| 217 |
+
"learning_rate": 4.131054131054131e-05,
|
| 218 |
+
"loss": 0.3105,
|
| 219 |
"step": 300
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.79,
|
| 223 |
+
"grad_norm": 4.1425909996032715,
|
| 224 |
+
"learning_rate": 4.083570750237417e-05,
|
| 225 |
+
"loss": 0.296,
|
| 226 |
"step": 310
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.82,
|
| 230 |
+
"grad_norm": 5.026642799377441,
|
| 231 |
+
"learning_rate": 4.036087369420703e-05,
|
| 232 |
+
"loss": 0.3594,
|
| 233 |
"step": 320
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.84,
|
| 237 |
+
"grad_norm": 3.949415922164917,
|
| 238 |
+
"learning_rate": 3.988603988603989e-05,
|
| 239 |
+
"loss": 0.2849,
|
| 240 |
"step": 330
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.87,
|
| 244 |
+
"grad_norm": 4.380434989929199,
|
| 245 |
+
"learning_rate": 3.941120607787275e-05,
|
| 246 |
+
"loss": 0.2935,
|
| 247 |
"step": 340
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.9,
|
| 251 |
+
"grad_norm": 4.884699821472168,
|
| 252 |
+
"learning_rate": 3.893637226970561e-05,
|
| 253 |
+
"loss": 0.3128,
|
| 254 |
"step": 350
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.92,
|
| 258 |
+
"grad_norm": 6.950473308563232,
|
| 259 |
+
"learning_rate": 3.846153846153846e-05,
|
| 260 |
+
"loss": 0.3073,
|
| 261 |
"step": 360
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.95,
|
| 265 |
+
"grad_norm": 5.410139560699463,
|
| 266 |
+
"learning_rate": 3.798670465337132e-05,
|
| 267 |
+
"loss": 0.2998,
|
| 268 |
"step": 370
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.97,
|
| 272 |
+
"grad_norm": 4.638174057006836,
|
| 273 |
+
"learning_rate": 3.7511870845204176e-05,
|
| 274 |
+
"loss": 0.3124,
|
| 275 |
"step": 380
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 1.0,
|
| 279 |
+
"grad_norm": 4.711712837219238,
|
| 280 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 281 |
+
"loss": 0.2682,
|
| 282 |
"step": 390
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 1.0,
|
| 286 |
+
"eval_accuracy": 0.9713,
|
| 287 |
+
"eval_loss": 0.0821285992860794,
|
| 288 |
+
"eval_runtime": 32.3489,
|
| 289 |
+
"eval_samples_per_second": 309.13,
|
| 290 |
+
"eval_steps_per_second": 9.676,
|
| 291 |
+
"step": 390
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.02,
|
| 295 |
+
"grad_norm": 4.979706764221191,
|
| 296 |
+
"learning_rate": 3.65622032288699e-05,
|
| 297 |
+
"loss": 0.2836,
|
| 298 |
"step": 400
|
| 299 |
},
|
| 300 |
{
|
| 301 |
+
"epoch": 1.05,
|
| 302 |
+
"grad_norm": 3.807408332824707,
|
| 303 |
+
"learning_rate": 3.608736942070276e-05,
|
| 304 |
+
"loss": 0.2521,
|
| 305 |
"step": 410
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 1.07,
|
| 309 |
+
"grad_norm": 4.402036190032959,
|
| 310 |
+
"learning_rate": 3.561253561253561e-05,
|
| 311 |
+
"loss": 0.2469,
|
| 312 |
"step": 420
|
| 313 |
},
|
| 314 |
{
|
| 315 |
+
"epoch": 1.1,
|
| 316 |
+
"grad_norm": 3.0667805671691895,
|
| 317 |
+
"learning_rate": 3.513770180436847e-05,
|
| 318 |
+
"loss": 0.2383,
|
| 319 |
"step": 430
|
| 320 |
},
|
| 321 |
{
|
| 322 |
+
"epoch": 1.13,
|
| 323 |
+
"grad_norm": 6.403406620025635,
|
| 324 |
+
"learning_rate": 3.466286799620133e-05,
|
| 325 |
+
"loss": 0.277,
|
| 326 |
"step": 440
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 1.15,
|
| 330 |
+
"grad_norm": 5.481479167938232,
|
| 331 |
+
"learning_rate": 3.418803418803419e-05,
|
| 332 |
+
"loss": 0.2533,
|
| 333 |
"step": 450
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 1.18,
|
| 337 |
+
"grad_norm": 5.324038028717041,
|
| 338 |
+
"learning_rate": 3.371320037986705e-05,
|
| 339 |
+
"loss": 0.2813,
|
| 340 |
"step": 460
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"epoch": 1.2,
|
| 344 |
+
"grad_norm": 2.4851675033569336,
|
| 345 |
+
"learning_rate": 3.323836657169991e-05,
|
| 346 |
+
"loss": 0.3055,
|
| 347 |
"step": 470
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 1.23,
|
| 351 |
+
"grad_norm": 4.365708827972412,
|
| 352 |
+
"learning_rate": 3.2763532763532764e-05,
|
| 353 |
+
"loss": 0.2797,
|
| 354 |
"step": 480
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 1.25,
|
| 358 |
+
"grad_norm": 3.6004509925842285,
|
| 359 |
+
"learning_rate": 3.2288698955365625e-05,
|
| 360 |
+
"loss": 0.2776,
|
| 361 |
"step": 490
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 1.28,
|
| 365 |
+
"grad_norm": 3.7064690589904785,
|
| 366 |
+
"learning_rate": 3.181386514719848e-05,
|
| 367 |
+
"loss": 0.2605,
|
| 368 |
"step": 500
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 1.31,
|
| 372 |
+
"grad_norm": 4.883594036102295,
|
| 373 |
+
"learning_rate": 3.133903133903134e-05,
|
| 374 |
+
"loss": 0.2403,
|
| 375 |
"step": 510
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 1.33,
|
| 379 |
+
"grad_norm": 5.014550685882568,
|
| 380 |
+
"learning_rate": 3.08641975308642e-05,
|
| 381 |
+
"loss": 0.2436,
|
| 382 |
"step": 520
|
| 383 |
},
|
| 384 |
{
|
| 385 |
+
"epoch": 1.36,
|
| 386 |
+
"grad_norm": 4.304734230041504,
|
| 387 |
+
"learning_rate": 3.0389363722697055e-05,
|
| 388 |
+
"loss": 0.2639,
|
| 389 |
"step": 530
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 1.38,
|
| 393 |
+
"grad_norm": 3.2759273052215576,
|
| 394 |
+
"learning_rate": 2.9914529914529915e-05,
|
| 395 |
+
"loss": 0.2651,
|
| 396 |
"step": 540
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 1.41,
|
| 400 |
+
"grad_norm": 3.556528091430664,
|
| 401 |
+
"learning_rate": 2.9439696106362775e-05,
|
| 402 |
+
"loss": 0.2359,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 403 |
"step": 550
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 1.43,
|
| 407 |
+
"grad_norm": 4.672804355621338,
|
| 408 |
+
"learning_rate": 2.8964862298195632e-05,
|
| 409 |
+
"loss": 0.2376,
|
| 410 |
"step": 560
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 1.46,
|
| 414 |
+
"grad_norm": 4.527768611907959,
|
| 415 |
+
"learning_rate": 2.8490028490028492e-05,
|
| 416 |
+
"loss": 0.2504,
|
| 417 |
"step": 570
|
| 418 |
},
|
| 419 |
{
|
| 420 |
+
"epoch": 1.48,
|
| 421 |
+
"grad_norm": 5.102674961090088,
|
| 422 |
+
"learning_rate": 2.8015194681861352e-05,
|
| 423 |
+
"loss": 0.2604,
|
| 424 |
"step": 580
|
| 425 |
},
|
| 426 |
{
|
| 427 |
+
"epoch": 1.51,
|
| 428 |
+
"grad_norm": 4.18637228012085,
|
| 429 |
+
"learning_rate": 2.754036087369421e-05,
|
| 430 |
+
"loss": 0.2746,
|
| 431 |
"step": 590
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 1.54,
|
| 435 |
+
"grad_norm": 3.3811545372009277,
|
| 436 |
+
"learning_rate": 2.706552706552707e-05,
|
| 437 |
+
"loss": 0.2548,
|
| 438 |
"step": 600
|
| 439 |
},
|
| 440 |
{
|
| 441 |
+
"epoch": 1.56,
|
| 442 |
+
"grad_norm": 3.1672608852386475,
|
| 443 |
+
"learning_rate": 2.6590693257359926e-05,
|
| 444 |
+
"loss": 0.2423,
|
| 445 |
"step": 610
|
| 446 |
},
|
| 447 |
{
|
| 448 |
+
"epoch": 1.59,
|
| 449 |
+
"grad_norm": 3.609534740447998,
|
| 450 |
+
"learning_rate": 2.611585944919278e-05,
|
| 451 |
+
"loss": 0.1994,
|
| 452 |
"step": 620
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 1.61,
|
| 456 |
+
"grad_norm": 5.083642482757568,
|
| 457 |
+
"learning_rate": 2.564102564102564e-05,
|
| 458 |
+
"loss": 0.292,
|
| 459 |
"step": 630
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 1.64,
|
| 463 |
+
"grad_norm": 4.716630935668945,
|
| 464 |
+
"learning_rate": 2.51661918328585e-05,
|
| 465 |
+
"loss": 0.2333,
|
| 466 |
"step": 640
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 1.66,
|
| 470 |
+
"grad_norm": 4.915971755981445,
|
| 471 |
+
"learning_rate": 2.4691358024691357e-05,
|
| 472 |
+
"loss": 0.2364,
|
| 473 |
"step": 650
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 1.69,
|
| 477 |
+
"grad_norm": 4.216696739196777,
|
| 478 |
+
"learning_rate": 2.4216524216524217e-05,
|
| 479 |
+
"loss": 0.2503,
|
| 480 |
"step": 660
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"epoch": 1.71,
|
| 484 |
+
"grad_norm": 4.966453552246094,
|
| 485 |
+
"learning_rate": 2.3741690408357077e-05,
|
| 486 |
+
"loss": 0.224,
|
| 487 |
"step": 670
|
| 488 |
},
|
| 489 |
{
|
| 490 |
+
"epoch": 1.74,
|
| 491 |
+
"grad_norm": 4.153652191162109,
|
| 492 |
+
"learning_rate": 2.3266856600189934e-05,
|
| 493 |
+
"loss": 0.2109,
|
| 494 |
"step": 680
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 1.77,
|
| 498 |
+
"grad_norm": 2.9919214248657227,
|
| 499 |
+
"learning_rate": 2.2792022792022794e-05,
|
| 500 |
+
"loss": 0.2371,
|
| 501 |
"step": 690
|
| 502 |
},
|
| 503 |
{
|
| 504 |
+
"epoch": 1.79,
|
| 505 |
+
"grad_norm": 5.105522155761719,
|
| 506 |
+
"learning_rate": 2.2317188983855654e-05,
|
| 507 |
+
"loss": 0.2187,
|
| 508 |
"step": 700
|
| 509 |
},
|
| 510 |
{
|
| 511 |
+
"epoch": 1.82,
|
| 512 |
+
"grad_norm": 3.9936702251434326,
|
| 513 |
+
"learning_rate": 2.184235517568851e-05,
|
| 514 |
+
"loss": 0.2403,
|
| 515 |
"step": 710
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"epoch": 1.84,
|
| 519 |
+
"grad_norm": 5.220417022705078,
|
| 520 |
+
"learning_rate": 2.1367521367521368e-05,
|
| 521 |
+
"loss": 0.2383,
|
| 522 |
"step": 720
|
| 523 |
},
|
| 524 |
{
|
| 525 |
+
"epoch": 1.87,
|
| 526 |
+
"grad_norm": 3.134110450744629,
|
| 527 |
+
"learning_rate": 2.0892687559354228e-05,
|
| 528 |
+
"loss": 0.2197,
|
| 529 |
"step": 730
|
| 530 |
},
|
| 531 |
{
|
| 532 |
+
"epoch": 1.89,
|
| 533 |
+
"grad_norm": 5.406284809112549,
|
| 534 |
+
"learning_rate": 2.0417853751187084e-05,
|
| 535 |
+
"loss": 0.224,
|
| 536 |
"step": 740
|
| 537 |
},
|
| 538 |
{
|
| 539 |
+
"epoch": 1.92,
|
| 540 |
+
"grad_norm": 5.273613452911377,
|
| 541 |
+
"learning_rate": 1.9943019943019945e-05,
|
| 542 |
+
"loss": 0.1951,
|
| 543 |
"step": 750
|
| 544 |
},
|
| 545 |
{
|
| 546 |
+
"epoch": 1.94,
|
| 547 |
+
"grad_norm": 4.695704936981201,
|
| 548 |
+
"learning_rate": 1.9468186134852805e-05,
|
| 549 |
+
"loss": 0.2324,
|
| 550 |
"step": 760
|
| 551 |
},
|
| 552 |
{
|
| 553 |
+
"epoch": 1.97,
|
| 554 |
+
"grad_norm": 5.734136581420898,
|
| 555 |
+
"learning_rate": 1.899335232668566e-05,
|
| 556 |
+
"loss": 0.248,
|
| 557 |
"step": 770
|
| 558 |
},
|
| 559 |
{
|
| 560 |
+
"epoch": 2.0,
|
| 561 |
+
"grad_norm": 4.757730960845947,
|
| 562 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 563 |
+
"loss": 0.252,
|
| 564 |
"step": 780
|
| 565 |
},
|
| 566 |
{
|
| 567 |
+
"epoch": 2.0,
|
| 568 |
+
"eval_accuracy": 0.9842,
|
| 569 |
+
"eval_loss": 0.050203289836645126,
|
| 570 |
+
"eval_runtime": 31.7681,
|
| 571 |
+
"eval_samples_per_second": 314.781,
|
| 572 |
+
"eval_steps_per_second": 9.853,
|
| 573 |
+
"step": 781
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 2.02,
|
| 577 |
+
"grad_norm": 4.400415897369385,
|
| 578 |
+
"learning_rate": 1.804368471035138e-05,
|
| 579 |
+
"loss": 0.2364,
|
| 580 |
"step": 790
|
| 581 |
},
|
| 582 |
{
|
| 583 |
+
"epoch": 2.05,
|
| 584 |
+
"grad_norm": 4.02318000793457,
|
| 585 |
+
"learning_rate": 1.7568850902184235e-05,
|
| 586 |
+
"loss": 0.2164,
|
| 587 |
"step": 800
|
| 588 |
},
|
| 589 |
{
|
| 590 |
+
"epoch": 2.07,
|
| 591 |
+
"grad_norm": 5.289691925048828,
|
| 592 |
+
"learning_rate": 1.7094017094017095e-05,
|
| 593 |
+
"loss": 0.2472,
|
| 594 |
"step": 810
|
| 595 |
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.1,
|
| 598 |
+
"grad_norm": 3.842559337615967,
|
| 599 |
+
"learning_rate": 1.6619183285849956e-05,
|
| 600 |
+
"loss": 0.2209,
|
| 601 |
+
"step": 820
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 2.12,
|
| 605 |
+
"grad_norm": 5.64124870300293,
|
| 606 |
+
"learning_rate": 1.6144349477682812e-05,
|
| 607 |
+
"loss": 0.2473,
|
| 608 |
+
"step": 830
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 2.15,
|
| 612 |
+
"grad_norm": 4.473005771636963,
|
| 613 |
+
"learning_rate": 1.566951566951567e-05,
|
| 614 |
+
"loss": 0.2287,
|
| 615 |
+
"step": 840
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 2.18,
|
| 619 |
+
"grad_norm": 4.476832389831543,
|
| 620 |
+
"learning_rate": 1.5194681861348528e-05,
|
| 621 |
+
"loss": 0.2208,
|
| 622 |
+
"step": 850
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.2,
|
| 626 |
+
"grad_norm": 4.89029598236084,
|
| 627 |
+
"learning_rate": 1.4719848053181388e-05,
|
| 628 |
+
"loss": 0.2332,
|
| 629 |
+
"step": 860
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 2.23,
|
| 633 |
+
"grad_norm": 5.0262627601623535,
|
| 634 |
+
"learning_rate": 1.4245014245014246e-05,
|
| 635 |
+
"loss": 0.2001,
|
| 636 |
+
"step": 870
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 2.25,
|
| 640 |
+
"grad_norm": 4.80778169631958,
|
| 641 |
+
"learning_rate": 1.3770180436847105e-05,
|
| 642 |
+
"loss": 0.2018,
|
| 643 |
+
"step": 880
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 2.28,
|
| 647 |
+
"grad_norm": 3.203700304031372,
|
| 648 |
+
"learning_rate": 1.3295346628679963e-05,
|
| 649 |
+
"loss": 0.2018,
|
| 650 |
+
"step": 890
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 2.3,
|
| 654 |
+
"grad_norm": 4.840007305145264,
|
| 655 |
+
"learning_rate": 1.282051282051282e-05,
|
| 656 |
+
"loss": 0.2754,
|
| 657 |
+
"step": 900
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 2.33,
|
| 661 |
+
"grad_norm": 4.501500129699707,
|
| 662 |
+
"learning_rate": 1.2345679012345678e-05,
|
| 663 |
+
"loss": 0.1986,
|
| 664 |
+
"step": 910
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 2.35,
|
| 668 |
+
"grad_norm": 3.4367220401763916,
|
| 669 |
+
"learning_rate": 1.1870845204178538e-05,
|
| 670 |
+
"loss": 0.2145,
|
| 671 |
+
"step": 920
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 2.38,
|
| 675 |
+
"grad_norm": 5.076148509979248,
|
| 676 |
+
"learning_rate": 1.1396011396011397e-05,
|
| 677 |
+
"loss": 0.2001,
|
| 678 |
+
"step": 930
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 2.41,
|
| 682 |
+
"grad_norm": 2.299694299697876,
|
| 683 |
+
"learning_rate": 1.0921177587844255e-05,
|
| 684 |
+
"loss": 0.1935,
|
| 685 |
+
"step": 940
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 2.43,
|
| 689 |
+
"grad_norm": 3.9149887561798096,
|
| 690 |
+
"learning_rate": 1.0446343779677114e-05,
|
| 691 |
+
"loss": 0.2158,
|
| 692 |
+
"step": 950
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 2.46,
|
| 696 |
+
"grad_norm": 5.087822437286377,
|
| 697 |
+
"learning_rate": 9.971509971509972e-06,
|
| 698 |
+
"loss": 0.2052,
|
| 699 |
+
"step": 960
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.48,
|
| 703 |
+
"grad_norm": 3.8664252758026123,
|
| 704 |
+
"learning_rate": 9.49667616334283e-06,
|
| 705 |
+
"loss": 0.2056,
|
| 706 |
+
"step": 970
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.51,
|
| 710 |
+
"grad_norm": 4.961308479309082,
|
| 711 |
+
"learning_rate": 9.02184235517569e-06,
|
| 712 |
+
"loss": 0.2157,
|
| 713 |
+
"step": 980
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 2.53,
|
| 717 |
+
"grad_norm": 4.553085803985596,
|
| 718 |
+
"learning_rate": 8.547008547008548e-06,
|
| 719 |
+
"loss": 0.2079,
|
| 720 |
+
"step": 990
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 2.56,
|
| 724 |
+
"grad_norm": 3.396073818206787,
|
| 725 |
+
"learning_rate": 8.072174738841406e-06,
|
| 726 |
+
"loss": 0.1919,
|
| 727 |
+
"step": 1000
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 2.58,
|
| 731 |
+
"grad_norm": 5.101813316345215,
|
| 732 |
+
"learning_rate": 7.597340930674264e-06,
|
| 733 |
+
"loss": 0.1812,
|
| 734 |
+
"step": 1010
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 2.61,
|
| 738 |
+
"grad_norm": 5.189543724060059,
|
| 739 |
+
"learning_rate": 7.122507122507123e-06,
|
| 740 |
+
"loss": 0.2123,
|
| 741 |
+
"step": 1020
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 2.64,
|
| 745 |
+
"grad_norm": 4.240951061248779,
|
| 746 |
+
"learning_rate": 6.6476733143399815e-06,
|
| 747 |
+
"loss": 0.218,
|
| 748 |
+
"step": 1030
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 2.66,
|
| 752 |
+
"grad_norm": 3.139678716659546,
|
| 753 |
+
"learning_rate": 6.172839506172839e-06,
|
| 754 |
+
"loss": 0.1824,
|
| 755 |
+
"step": 1040
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 2.69,
|
| 759 |
+
"grad_norm": 2.84495210647583,
|
| 760 |
+
"learning_rate": 5.6980056980056985e-06,
|
| 761 |
+
"loss": 0.1916,
|
| 762 |
+
"step": 1050
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 2.71,
|
| 766 |
+
"grad_norm": 3.4231679439544678,
|
| 767 |
+
"learning_rate": 5.223171889838557e-06,
|
| 768 |
+
"loss": 0.1953,
|
| 769 |
+
"step": 1060
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 2.74,
|
| 773 |
+
"grad_norm": 4.120250225067139,
|
| 774 |
+
"learning_rate": 4.748338081671415e-06,
|
| 775 |
+
"loss": 0.2046,
|
| 776 |
+
"step": 1070
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 2.76,
|
| 780 |
+
"grad_norm": 5.0515055656433105,
|
| 781 |
+
"learning_rate": 4.273504273504274e-06,
|
| 782 |
+
"loss": 0.1941,
|
| 783 |
+
"step": 1080
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 2.79,
|
| 787 |
+
"grad_norm": 4.197494029998779,
|
| 788 |
+
"learning_rate": 3.798670465337132e-06,
|
| 789 |
+
"loss": 0.1908,
|
| 790 |
+
"step": 1090
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 2.82,
|
| 794 |
+
"grad_norm": 5.316411972045898,
|
| 795 |
+
"learning_rate": 3.3238366571699908e-06,
|
| 796 |
+
"loss": 0.1954,
|
| 797 |
+
"step": 1100
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 2.84,
|
| 801 |
+
"grad_norm": 2.8527472019195557,
|
| 802 |
+
"learning_rate": 2.8490028490028492e-06,
|
| 803 |
+
"loss": 0.1466,
|
| 804 |
+
"step": 1110
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 2.87,
|
| 808 |
+
"grad_norm": 4.074756622314453,
|
| 809 |
+
"learning_rate": 2.3741690408357077e-06,
|
| 810 |
+
"loss": 0.2092,
|
| 811 |
+
"step": 1120
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 2.89,
|
| 815 |
+
"grad_norm": 4.336794853210449,
|
| 816 |
+
"learning_rate": 1.899335232668566e-06,
|
| 817 |
+
"loss": 0.1742,
|
| 818 |
+
"step": 1130
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 2.92,
|
| 822 |
+
"grad_norm": 3.8528594970703125,
|
| 823 |
+
"learning_rate": 1.4245014245014246e-06,
|
| 824 |
+
"loss": 0.2002,
|
| 825 |
+
"step": 1140
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 2.94,
|
| 829 |
+
"grad_norm": 4.159782409667969,
|
| 830 |
+
"learning_rate": 9.49667616334283e-07,
|
| 831 |
+
"loss": 0.1847,
|
| 832 |
+
"step": 1150
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 2.97,
|
| 836 |
+
"grad_norm": 5.317773342132568,
|
| 837 |
+
"learning_rate": 4.748338081671415e-07,
|
| 838 |
+
"loss": 0.2293,
|
| 839 |
+
"step": 1160
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 2.99,
|
| 843 |
+
"grad_norm": 4.241806983947754,
|
| 844 |
+
"learning_rate": 0.0,
|
| 845 |
+
"loss": 0.173,
|
| 846 |
+
"step": 1170
|
| 847 |
+
},
|
| 848 |
{
|
| 849 |
"epoch": 2.99,
|
| 850 |
+
"eval_accuracy": 0.9844,
|
| 851 |
+
"eval_loss": 0.04590694606304169,
|
| 852 |
+
"eval_runtime": 32.5039,
|
| 853 |
+
"eval_samples_per_second": 307.655,
|
| 854 |
+
"eval_steps_per_second": 9.63,
|
| 855 |
+
"step": 1170
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 2.99,
|
| 859 |
+
"step": 1170,
|
| 860 |
+
"total_flos": 2.930358373492064e+18,
|
| 861 |
+
"train_loss": 0.33610059461023056,
|
| 862 |
+
"train_runtime": 935.657,
|
| 863 |
+
"train_samples_per_second": 160.315,
|
| 864 |
+
"train_steps_per_second": 1.25
|
| 865 |
}
|
| 866 |
],
|
| 867 |
"logging_steps": 10,
|
| 868 |
+
"max_steps": 1170,
|
| 869 |
"num_input_tokens_seen": 0,
|
| 870 |
"num_train_epochs": 3,
|
| 871 |
"save_steps": 500,
|
| 872 |
+
"total_flos": 2.930358373492064e+18,
|
| 873 |
"train_batch_size": 32,
|
| 874 |
"trial_name": null,
|
| 875 |
"trial_params": null
|