Commit
•
8aeb7d8
1
Parent(s):
5cf2b02
End of training
Browse files- README.md +1 -1
- all_results.json +10 -10
- eval_results.json +6 -6
- runs/May10_16-51-32_5ef4b5b162fa/events.out.tfevents.1715368283.5ef4b5b162fa.34.4 +3 -0
- train_results.json +5 -5
- trainer_state.json +760 -365
README.md
CHANGED
|
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
- Accuracy: 0.9264
|
| 22 |
|
| 23 |
## Model description
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.2787
|
| 21 |
- Accuracy: 0.9264
|
| 22 |
|
| 23 |
## Model description
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 71.
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second": 52.
|
| 12 |
"train_steps_per_second": 0.204
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 14.96,
|
| 3 |
+
"eval_accuracy": 0.926367461430575,
|
| 4 |
+
"eval_loss": 0.27871325612068176,
|
| 5 |
+
"eval_runtime": 40.1195,
|
| 6 |
+
"eval_samples_per_second": 71.088,
|
| 7 |
+
"eval_steps_per_second": 1.122,
|
| 8 |
+
"total_flos": 9.546299981955072e+18,
|
| 9 |
+
"train_loss": 0.11206140112876892,
|
| 10 |
+
"train_runtime": 7336.5108,
|
| 11 |
+
"train_samples_per_second": 52.472,
|
| 12 |
"train_steps_per_second": 0.204
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 71.
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 14.96,
|
| 3 |
+
"eval_accuracy": 0.926367461430575,
|
| 4 |
+
"eval_loss": 0.27871325612068176,
|
| 5 |
+
"eval_runtime": 40.1195,
|
| 6 |
+
"eval_samples_per_second": 71.088,
|
| 7 |
+
"eval_steps_per_second": 1.122
|
| 8 |
}
|
runs/May10_16-51-32_5ef4b5b162fa/events.out.tfevents.1715368283.5ef4b5b162fa.34.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f8bdf8803bb3d978ec3d8d00dec1a3e59255f8131a3b1f83dd4d8dcf7228a9a
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second": 52.
|
| 7 |
"train_steps_per_second": 0.204
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 14.96,
|
| 3 |
+
"total_flos": 9.546299981955072e+18,
|
| 4 |
+
"train_loss": 0.11206140112876892,
|
| 5 |
+
"train_runtime": 7336.5108,
|
| 6 |
+
"train_samples_per_second": 52.472,
|
| 7 |
"train_steps_per_second": 0.204
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,819 +1,1214 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-classification/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.1,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.2,
|
| 20 |
-
"grad_norm": 4.
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.3,
|
| 27 |
-
"grad_norm": 4.
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.4,
|
| 34 |
-
"grad_norm": 4.
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.5,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.6,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.7,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.8,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.9,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 1.0,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.0,
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_loss": 0.
|
| 85 |
-
"eval_runtime":
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second": 1.
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.1,
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.2,
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate":
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 1.3,
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate":
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 1.4,
|
| 113 |
-
"grad_norm": 6.
|
| 114 |
-
"learning_rate": 9.
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 1.5,
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 1.6,
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate": 9.
|
| 129 |
-
"loss": 0.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 1.7,
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate": 9.
|
| 136 |
-
"loss": 0.
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.8,
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate": 9.
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 1.9,
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"learning_rate":
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 2.0,
|
| 155 |
-
"grad_norm":
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 2.0,
|
| 162 |
-
"eval_accuracy": 0.
|
| 163 |
-
"eval_loss": 0.
|
| 164 |
-
"eval_runtime":
|
| 165 |
-
"eval_samples_per_second":
|
| 166 |
-
"eval_steps_per_second": 1.
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 2.09,
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 2.19,
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate":
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 2.29,
|
| 185 |
-
"grad_norm":
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 2.39,
|
| 192 |
-
"grad_norm": 5.
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 2.49,
|
| 199 |
-
"grad_norm":
|
| 200 |
-
"learning_rate":
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 2.59,
|
| 206 |
-
"grad_norm": 7.
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 2.69,
|
| 213 |
-
"grad_norm": 5.
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 2.79,
|
| 220 |
-
"grad_norm": 5.
|
| 221 |
-
"learning_rate":
|
| 222 |
-
"loss": 0.
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 2.89,
|
| 227 |
-
"grad_norm": 5.
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 2.99,
|
| 234 |
-
"grad_norm": 4.
|
| 235 |
-
"learning_rate":
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 2.99,
|
| 241 |
-
"eval_accuracy": 0.
|
| 242 |
-
"eval_loss": 0.
|
| 243 |
-
"eval_runtime": 39.
|
| 244 |
-
"eval_samples_per_second": 72.
|
| 245 |
-
"eval_steps_per_second": 1.
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 3.09,
|
| 250 |
-
"grad_norm":
|
| 251 |
-
"learning_rate":
|
| 252 |
-
"loss": 0.
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 3.19,
|
| 257 |
-
"grad_norm":
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 3.29,
|
| 264 |
-
"grad_norm":
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 3.39,
|
| 271 |
-
"grad_norm":
|
| 272 |
-
"learning_rate":
|
| 273 |
-
"loss": 0.
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 3.49,
|
| 278 |
-
"grad_norm":
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 3.59,
|
| 285 |
-
"grad_norm":
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 3.69,
|
| 292 |
-
"grad_norm":
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 3.79,
|
| 299 |
-
"grad_norm":
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 3.89,
|
| 306 |
-
"grad_norm":
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 3.99,
|
| 313 |
-
"grad_norm":
|
| 314 |
-
"learning_rate":
|
| 315 |
-
"loss": 0.
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 4.0,
|
| 320 |
-
"eval_accuracy": 0.
|
| 321 |
-
"eval_loss": 0.
|
| 322 |
-
"eval_runtime": 39.
|
| 323 |
-
"eval_samples_per_second": 72.
|
| 324 |
-
"eval_steps_per_second": 1.
|
| 325 |
"step": 401
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 4.09,
|
| 329 |
-
"grad_norm": 5.
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 4.19,
|
| 336 |
-
"grad_norm":
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 4.29,
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 4.39,
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 4.49,
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 4.59,
|
| 364 |
-
"grad_norm":
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 4.69,
|
| 371 |
-
"grad_norm":
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 4.79,
|
| 378 |
-
"grad_norm": 4.
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 4.89,
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 4.99,
|
| 392 |
-
"grad_norm": 4.
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 5.0,
|
| 399 |
-
"eval_accuracy": 0.
|
| 400 |
-
"eval_loss": 0.
|
| 401 |
-
"eval_runtime": 39.
|
| 402 |
-
"eval_samples_per_second":
|
| 403 |
-
"eval_steps_per_second": 1.
|
| 404 |
"step": 501
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 5.09,
|
| 408 |
-
"grad_norm":
|
| 409 |
-
"learning_rate":
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 5.19,
|
| 415 |
-
"grad_norm":
|
| 416 |
-
"learning_rate":
|
| 417 |
-
"loss": 0.
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 5.29,
|
| 422 |
-
"grad_norm": 4.
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 5.39,
|
| 429 |
-
"grad_norm":
|
| 430 |
-
"learning_rate":
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 5.49,
|
| 436 |
-
"grad_norm":
|
| 437 |
-
"learning_rate":
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 5.59,
|
| 443 |
-
"grad_norm":
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 5.69,
|
| 450 |
-
"grad_norm":
|
| 451 |
-
"learning_rate":
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 5.79,
|
| 457 |
-
"grad_norm":
|
| 458 |
-
"learning_rate":
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 5.89,
|
| 464 |
-
"grad_norm":
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 5.99,
|
| 471 |
-
"grad_norm": 3.
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 6.0,
|
| 478 |
-
"eval_accuracy": 0.
|
| 479 |
-
"eval_loss": 0.
|
| 480 |
-
"eval_runtime":
|
| 481 |
-
"eval_samples_per_second":
|
| 482 |
-
"eval_steps_per_second": 1.
|
| 483 |
"step": 601
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 6.08,
|
| 487 |
-
"grad_norm":
|
| 488 |
-
"learning_rate":
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 6.18,
|
| 494 |
-
"grad_norm":
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 6.28,
|
| 501 |
-
"grad_norm":
|
| 502 |
-
"learning_rate":
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 6.38,
|
| 508 |
-
"grad_norm": 4.
|
| 509 |
-
"learning_rate":
|
| 510 |
-
"loss": 0.
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 6.48,
|
| 515 |
-
"grad_norm":
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 6.58,
|
| 522 |
-
"grad_norm": 4.
|
| 523 |
-
"learning_rate":
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 6.68,
|
| 529 |
-
"grad_norm": 4.
|
| 530 |
-
"learning_rate":
|
| 531 |
-
"loss": 0.
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 6.78,
|
| 536 |
-
"grad_norm": 5.
|
| 537 |
-
"learning_rate":
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 6.88,
|
| 543 |
-
"grad_norm": 3.
|
| 544 |
-
"learning_rate":
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 6.98,
|
| 550 |
-
"grad_norm":
|
| 551 |
-
"learning_rate":
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 6.99,
|
| 557 |
-
"eval_accuracy": 0.
|
| 558 |
-
"eval_loss": 0.
|
| 559 |
-
"eval_runtime": 39.
|
| 560 |
-
"eval_samples_per_second": 72.
|
| 561 |
-
"eval_steps_per_second": 1.
|
| 562 |
"step": 701
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 7.08,
|
| 566 |
-
"grad_norm":
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 7.18,
|
| 573 |
-
"grad_norm": 3.
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 7.28,
|
| 580 |
-
"grad_norm":
|
| 581 |
-
"learning_rate":
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 7.38,
|
| 587 |
-
"grad_norm":
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 7.48,
|
| 594 |
-
"grad_norm":
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 7.58,
|
| 601 |
-
"grad_norm":
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 7.68,
|
| 608 |
-
"grad_norm":
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 7.78,
|
| 615 |
-
"grad_norm":
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 7.88,
|
| 622 |
-
"grad_norm":
|
| 623 |
-
"learning_rate":
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 7.98,
|
| 629 |
-
"grad_norm": 4.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 8.0,
|
| 636 |
-
"eval_accuracy": 0.
|
| 637 |
-
"eval_loss": 0.
|
| 638 |
-
"eval_runtime": 39.
|
| 639 |
-
"eval_samples_per_second":
|
| 640 |
-
"eval_steps_per_second": 1.
|
| 641 |
"step": 802
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 8.08,
|
| 645 |
-
"grad_norm": 3.
|
| 646 |
-
"learning_rate":
|
| 647 |
-
"loss": 0.
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
"epoch": 8.18,
|
| 652 |
-
"grad_norm": 4.
|
| 653 |
-
"learning_rate":
|
| 654 |
-
"loss": 0.
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 8.28,
|
| 659 |
-
"grad_norm":
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 8.38,
|
| 666 |
-
"grad_norm":
|
| 667 |
-
"learning_rate":
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 8.48,
|
| 673 |
-
"grad_norm":
|
| 674 |
-
"learning_rate":
|
| 675 |
-
"loss": 0.
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 8.58,
|
| 680 |
-
"grad_norm": 3.
|
| 681 |
-
"learning_rate":
|
| 682 |
-
"loss": 0.
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 8.68,
|
| 687 |
-
"grad_norm": 4.
|
| 688 |
-
"learning_rate":
|
| 689 |
-
"loss": 0.
|
| 690 |
"step": 870
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 8.78,
|
| 694 |
-
"grad_norm":
|
| 695 |
-
"learning_rate":
|
| 696 |
-
"loss": 0.
|
| 697 |
"step": 880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 8.88,
|
| 701 |
-
"grad_norm":
|
| 702 |
-
"learning_rate":
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 890
|
| 705 |
},
|
| 706 |
{
|
| 707 |
"epoch": 8.98,
|
| 708 |
-
"grad_norm":
|
| 709 |
-
"learning_rate":
|
| 710 |
-
"loss": 0.
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 9.0,
|
| 715 |
-
"eval_accuracy": 0.
|
| 716 |
-
"eval_loss": 0.
|
| 717 |
-
"eval_runtime": 39.
|
| 718 |
-
"eval_samples_per_second":
|
| 719 |
-
"eval_steps_per_second": 1.
|
| 720 |
"step": 902
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 9.08,
|
| 724 |
-
"grad_norm": 4.
|
| 725 |
-
"learning_rate":
|
| 726 |
-
"loss": 0.
|
| 727 |
"step": 910
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 9.18,
|
| 731 |
-
"grad_norm": 4.
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 920
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 9.28,
|
| 738 |
-
"grad_norm":
|
| 739 |
-
"learning_rate":
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 930
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 9.38,
|
| 745 |
-
"grad_norm": 4.
|
| 746 |
-
"learning_rate":
|
| 747 |
-
"loss": 0.
|
| 748 |
"step": 940
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 9.48,
|
| 752 |
-
"grad_norm":
|
| 753 |
-
"learning_rate":
|
| 754 |
-
"loss": 0.
|
| 755 |
"step": 950
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 9.58,
|
| 759 |
-
"grad_norm":
|
| 760 |
-
"learning_rate":
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 960
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 9.68,
|
| 766 |
-
"grad_norm":
|
| 767 |
-
"learning_rate": 3.
|
| 768 |
-
"loss": 0.
|
| 769 |
"step": 970
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 9.78,
|
| 773 |
-
"grad_norm": 3.
|
| 774 |
-
"learning_rate":
|
| 775 |
-
"loss": 0.
|
| 776 |
"step": 980
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 9.88,
|
| 780 |
-
"grad_norm":
|
| 781 |
-
"learning_rate":
|
| 782 |
-
"loss": 0.
|
| 783 |
"step": 990
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 9.98,
|
| 787 |
-
"grad_norm":
|
| 788 |
-
"learning_rate":
|
| 789 |
-
"loss": 0.
|
| 790 |
"step": 1000
|
| 791 |
},
|
| 792 |
{
|
| 793 |
-
"epoch":
|
| 794 |
-
"eval_accuracy": 0.
|
| 795 |
-
"eval_loss": 0.
|
| 796 |
-
"eval_runtime": 39.
|
| 797 |
-
"eval_samples_per_second": 72.
|
| 798 |
-
"eval_steps_per_second": 1.
|
| 799 |
-
"step":
|
| 800 |
},
|
| 801 |
{
|
| 802 |
-
"epoch":
|
| 803 |
-
"
|
| 804 |
-
"
|
| 805 |
-
"
|
| 806 |
-
"
|
| 807 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 808 |
"train_steps_per_second": 0.204
|
| 809 |
}
|
| 810 |
],
|
| 811 |
"logging_steps": 10,
|
| 812 |
-
"max_steps":
|
| 813 |
"num_input_tokens_seen": 0,
|
| 814 |
-
"num_train_epochs":
|
| 815 |
"save_steps": 500,
|
| 816 |
-
"total_flos":
|
| 817 |
"train_batch_size": 64,
|
| 818 |
"trial_name": null,
|
| 819 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.926367461430575,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-classification/checkpoint-1403",
|
| 4 |
+
"epoch": 14.962593516209477,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1500,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.1,
|
| 13 |
+
"grad_norm": 3.234764814376831,
|
| 14 |
+
"learning_rate": 6.666666666666667e-06,
|
| 15 |
+
"loss": 0.1047,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.2,
|
| 20 |
+
"grad_norm": 4.926308631896973,
|
| 21 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 22 |
+
"loss": 0.1257,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.3,
|
| 27 |
+
"grad_norm": 4.245235919952393,
|
| 28 |
+
"learning_rate": 2e-05,
|
| 29 |
+
"loss": 0.1047,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.4,
|
| 34 |
+
"grad_norm": 4.806109428405762,
|
| 35 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 36 |
+
"loss": 0.1283,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.5,
|
| 41 |
+
"grad_norm": 4.090646743774414,
|
| 42 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 43 |
+
"loss": 0.1193,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.6,
|
| 48 |
+
"grad_norm": 3.47908353805542,
|
| 49 |
+
"learning_rate": 4e-05,
|
| 50 |
+
"loss": 0.1233,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.7,
|
| 55 |
+
"grad_norm": 3.029428720474243,
|
| 56 |
+
"learning_rate": 4.666666666666667e-05,
|
| 57 |
+
"loss": 0.1364,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.8,
|
| 62 |
+
"grad_norm": 5.012099742889404,
|
| 63 |
+
"learning_rate": 5.333333333333333e-05,
|
| 64 |
+
"loss": 0.1371,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.9,
|
| 69 |
+
"grad_norm": 4.080081462860107,
|
| 70 |
+
"learning_rate": 6e-05,
|
| 71 |
+
"loss": 0.1219,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 1.0,
|
| 76 |
+
"grad_norm": 4.989726543426514,
|
| 77 |
+
"learning_rate": 6.666666666666667e-05,
|
| 78 |
+
"loss": 0.1469,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.0,
|
| 83 |
+
"eval_accuracy": 0.9126928471248247,
|
| 84 |
+
"eval_loss": 0.302682101726532,
|
| 85 |
+
"eval_runtime": 38.9184,
|
| 86 |
+
"eval_samples_per_second": 73.282,
|
| 87 |
+
"eval_steps_per_second": 1.156,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.1,
|
| 92 |
+
"grad_norm": 4.7255024909973145,
|
| 93 |
+
"learning_rate": 7.333333333333333e-05,
|
| 94 |
+
"loss": 0.1329,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.2,
|
| 99 |
+
"grad_norm": 7.446033477783203,
|
| 100 |
+
"learning_rate": 8e-05,
|
| 101 |
+
"loss": 0.1485,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 1.3,
|
| 106 |
+
"grad_norm": 7.688333988189697,
|
| 107 |
+
"learning_rate": 8.666666666666667e-05,
|
| 108 |
+
"loss": 0.1558,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 1.4,
|
| 113 |
+
"grad_norm": 6.560573577880859,
|
| 114 |
+
"learning_rate": 9.333333333333334e-05,
|
| 115 |
+
"loss": 0.1516,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 1.5,
|
| 120 |
+
"grad_norm": 4.632976055145264,
|
| 121 |
+
"learning_rate": 0.0001,
|
| 122 |
+
"loss": 0.153,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 1.6,
|
| 127 |
+
"grad_norm": 6.091333389282227,
|
| 128 |
+
"learning_rate": 9.925925925925926e-05,
|
| 129 |
+
"loss": 0.1659,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 1.7,
|
| 134 |
+
"grad_norm": 4.933795928955078,
|
| 135 |
+
"learning_rate": 9.851851851851852e-05,
|
| 136 |
+
"loss": 0.1837,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.8,
|
| 141 |
+
"grad_norm": 7.728570938110352,
|
| 142 |
+
"learning_rate": 9.777777777777778e-05,
|
| 143 |
+
"loss": 0.1817,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 1.9,
|
| 148 |
+
"grad_norm": 5.771063327789307,
|
| 149 |
+
"learning_rate": 9.703703703703704e-05,
|
| 150 |
+
"loss": 0.1667,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 2.0,
|
| 155 |
+
"grad_norm": 4.893702030181885,
|
| 156 |
+
"learning_rate": 9.62962962962963e-05,
|
| 157 |
+
"loss": 0.1677,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 2.0,
|
| 162 |
+
"eval_accuracy": 0.9000701262272089,
|
| 163 |
+
"eval_loss": 0.335124135017395,
|
| 164 |
+
"eval_runtime": 39.1392,
|
| 165 |
+
"eval_samples_per_second": 72.868,
|
| 166 |
+
"eval_steps_per_second": 1.15,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 2.09,
|
| 171 |
+
"grad_norm": 5.826776027679443,
|
| 172 |
+
"learning_rate": 9.555555555555557e-05,
|
| 173 |
+
"loss": 0.1539,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 2.19,
|
| 178 |
+
"grad_norm": 3.6135647296905518,
|
| 179 |
+
"learning_rate": 9.481481481481483e-05,
|
| 180 |
+
"loss": 0.144,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 2.29,
|
| 185 |
+
"grad_norm": 4.992943286895752,
|
| 186 |
+
"learning_rate": 9.407407407407408e-05,
|
| 187 |
+
"loss": 0.1626,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 2.39,
|
| 192 |
+
"grad_norm": 5.117826461791992,
|
| 193 |
+
"learning_rate": 9.333333333333334e-05,
|
| 194 |
+
"loss": 0.1459,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 2.49,
|
| 199 |
+
"grad_norm": 5.909987926483154,
|
| 200 |
+
"learning_rate": 9.25925925925926e-05,
|
| 201 |
+
"loss": 0.1577,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 2.59,
|
| 206 |
+
"grad_norm": 7.713306903839111,
|
| 207 |
+
"learning_rate": 9.185185185185186e-05,
|
| 208 |
+
"loss": 0.1829,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 2.69,
|
| 213 |
+
"grad_norm": 5.385562419891357,
|
| 214 |
+
"learning_rate": 9.111111111111112e-05,
|
| 215 |
+
"loss": 0.1298,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 2.79,
|
| 220 |
+
"grad_norm": 5.207605838775635,
|
| 221 |
+
"learning_rate": 9.037037037037038e-05,
|
| 222 |
+
"loss": 0.1794,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 2.89,
|
| 227 |
+
"grad_norm": 5.401705265045166,
|
| 228 |
+
"learning_rate": 8.962962962962963e-05,
|
| 229 |
+
"loss": 0.1632,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 2.99,
|
| 234 |
+
"grad_norm": 4.565079212188721,
|
| 235 |
+
"learning_rate": 8.888888888888889e-05,
|
| 236 |
+
"loss": 0.167,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 2.99,
|
| 241 |
+
"eval_accuracy": 0.8930575035063114,
|
| 242 |
+
"eval_loss": 0.3874761760234833,
|
| 243 |
+
"eval_runtime": 39.183,
|
| 244 |
+
"eval_samples_per_second": 72.787,
|
| 245 |
+
"eval_steps_per_second": 1.148,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 3.09,
|
| 250 |
+
"grad_norm": 6.44505500793457,
|
| 251 |
+
"learning_rate": 8.814814814814815e-05,
|
| 252 |
+
"loss": 0.1483,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 3.19,
|
| 257 |
+
"grad_norm": 7.901595592498779,
|
| 258 |
+
"learning_rate": 8.740740740740741e-05,
|
| 259 |
+
"loss": 0.1422,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 3.29,
|
| 264 |
+
"grad_norm": 6.531421184539795,
|
| 265 |
+
"learning_rate": 8.666666666666667e-05,
|
| 266 |
+
"loss": 0.145,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 3.39,
|
| 271 |
+
"grad_norm": 3.7486908435821533,
|
| 272 |
+
"learning_rate": 8.592592592592593e-05,
|
| 273 |
+
"loss": 0.1519,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 3.49,
|
| 278 |
+
"grad_norm": 4.0806403160095215,
|
| 279 |
+
"learning_rate": 8.518518518518518e-05,
|
| 280 |
+
"loss": 0.1519,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 3.59,
|
| 285 |
+
"grad_norm": 4.170175552368164,
|
| 286 |
+
"learning_rate": 8.444444444444444e-05,
|
| 287 |
+
"loss": 0.1287,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 3.69,
|
| 292 |
+
"grad_norm": 3.4781312942504883,
|
| 293 |
+
"learning_rate": 8.37037037037037e-05,
|
| 294 |
+
"loss": 0.1338,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 3.79,
|
| 299 |
+
"grad_norm": 5.322227478027344,
|
| 300 |
+
"learning_rate": 8.296296296296296e-05,
|
| 301 |
+
"loss": 0.1653,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 3.89,
|
| 306 |
+
"grad_norm": 6.252974510192871,
|
| 307 |
+
"learning_rate": 8.222222222222222e-05,
|
| 308 |
+
"loss": 0.1527,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 3.99,
|
| 313 |
+
"grad_norm": 3.7561535835266113,
|
| 314 |
+
"learning_rate": 8.148148148148148e-05,
|
| 315 |
+
"loss": 0.1556,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 4.0,
|
| 320 |
+
"eval_accuracy": 0.896914446002805,
|
| 321 |
+
"eval_loss": 0.3814311921596527,
|
| 322 |
+
"eval_runtime": 39.4669,
|
| 323 |
+
"eval_samples_per_second": 72.263,
|
| 324 |
+
"eval_steps_per_second": 1.14,
|
| 325 |
"step": 401
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 4.09,
|
| 329 |
+
"grad_norm": 5.396999359130859,
|
| 330 |
+
"learning_rate": 8.074074074074075e-05,
|
| 331 |
+
"loss": 0.1274,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 4.19,
|
| 336 |
+
"grad_norm": 4.547370910644531,
|
| 337 |
+
"learning_rate": 8e-05,
|
| 338 |
+
"loss": 0.1089,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 4.29,
|
| 343 |
+
"grad_norm": 6.839280605316162,
|
| 344 |
+
"learning_rate": 7.925925925925926e-05,
|
| 345 |
+
"loss": 0.1402,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 4.39,
|
| 350 |
+
"grad_norm": 8.286632537841797,
|
| 351 |
+
"learning_rate": 7.851851851851852e-05,
|
| 352 |
+
"loss": 0.1313,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 4.49,
|
| 357 |
+
"grad_norm": 3.8082900047302246,
|
| 358 |
+
"learning_rate": 7.777777777777778e-05,
|
| 359 |
+
"loss": 0.1231,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 4.59,
|
| 364 |
+
"grad_norm": 6.801741123199463,
|
| 365 |
+
"learning_rate": 7.703703703703704e-05,
|
| 366 |
+
"loss": 0.1387,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 4.69,
|
| 371 |
+
"grad_norm": 4.6010823249816895,
|
| 372 |
+
"learning_rate": 7.62962962962963e-05,
|
| 373 |
+
"loss": 0.1273,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 4.79,
|
| 378 |
+
"grad_norm": 4.252594947814941,
|
| 379 |
+
"learning_rate": 7.555555555555556e-05,
|
| 380 |
+
"loss": 0.1276,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 4.89,
|
| 385 |
+
"grad_norm": 4.800199031829834,
|
| 386 |
+
"learning_rate": 7.481481481481481e-05,
|
| 387 |
+
"loss": 0.1282,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 4.99,
|
| 392 |
+
"grad_norm": 4.848538875579834,
|
| 393 |
+
"learning_rate": 7.407407407407407e-05,
|
| 394 |
+
"loss": 0.1328,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 5.0,
|
| 399 |
+
"eval_accuracy": 0.9046283309957924,
|
| 400 |
+
"eval_loss": 0.32806462049484253,
|
| 401 |
+
"eval_runtime": 39.0575,
|
| 402 |
+
"eval_samples_per_second": 73.02,
|
| 403 |
+
"eval_steps_per_second": 1.152,
|
| 404 |
"step": 501
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 5.09,
|
| 408 |
+
"grad_norm": 5.300040245056152,
|
| 409 |
+
"learning_rate": 7.333333333333333e-05,
|
| 410 |
+
"loss": 0.1063,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 5.19,
|
| 415 |
+
"grad_norm": 4.112715721130371,
|
| 416 |
+
"learning_rate": 7.25925925925926e-05,
|
| 417 |
+
"loss": 0.1425,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 5.29,
|
| 422 |
+
"grad_norm": 4.367955684661865,
|
| 423 |
+
"learning_rate": 7.185185185185186e-05,
|
| 424 |
+
"loss": 0.1198,
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 5.39,
|
| 429 |
+
"grad_norm": 3.767223834991455,
|
| 430 |
+
"learning_rate": 7.111111111111112e-05,
|
| 431 |
+
"loss": 0.1199,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 5.49,
|
| 436 |
+
"grad_norm": 4.640177249908447,
|
| 437 |
+
"learning_rate": 7.037037037037038e-05,
|
| 438 |
+
"loss": 0.1175,
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 5.59,
|
| 443 |
+
"grad_norm": 4.267022609710693,
|
| 444 |
+
"learning_rate": 6.962962962962964e-05,
|
| 445 |
+
"loss": 0.1312,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 5.69,
|
| 450 |
+
"grad_norm": 3.086017608642578,
|
| 451 |
+
"learning_rate": 6.88888888888889e-05,
|
| 452 |
+
"loss": 0.1055,
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 5.79,
|
| 457 |
+
"grad_norm": 6.53023099899292,
|
| 458 |
+
"learning_rate": 6.814814814814815e-05,
|
| 459 |
+
"loss": 0.126,
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 5.89,
|
| 464 |
+
"grad_norm": 3.820864677429199,
|
| 465 |
+
"learning_rate": 6.740740740740741e-05,
|
| 466 |
+
"loss": 0.1197,
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 5.99,
|
| 471 |
+
"grad_norm": 3.039929151535034,
|
| 472 |
+
"learning_rate": 6.666666666666667e-05,
|
| 473 |
+
"loss": 0.1,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 6.0,
|
| 478 |
+
"eval_accuracy": 0.9004207573632539,
|
| 479 |
+
"eval_loss": 0.3726123869419098,
|
| 480 |
+
"eval_runtime": 39.249,
|
| 481 |
+
"eval_samples_per_second": 72.664,
|
| 482 |
+
"eval_steps_per_second": 1.147,
|
| 483 |
"step": 601
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 6.08,
|
| 487 |
+
"grad_norm": 5.862295150756836,
|
| 488 |
+
"learning_rate": 6.592592592592593e-05,
|
| 489 |
+
"loss": 0.1248,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 6.18,
|
| 494 |
+
"grad_norm": 3.471050977706909,
|
| 495 |
+
"learning_rate": 6.51851851851852e-05,
|
| 496 |
+
"loss": 0.1107,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 6.28,
|
| 501 |
+
"grad_norm": 3.645636558532715,
|
| 502 |
+
"learning_rate": 6.444444444444446e-05,
|
| 503 |
+
"loss": 0.0999,
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 6.38,
|
| 508 |
+
"grad_norm": 4.186192512512207,
|
| 509 |
+
"learning_rate": 6.37037037037037e-05,
|
| 510 |
+
"loss": 0.1281,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 6.48,
|
| 515 |
+
"grad_norm": 5.014218807220459,
|
| 516 |
+
"learning_rate": 6.296296296296296e-05,
|
| 517 |
+
"loss": 0.115,
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 6.58,
|
| 522 |
+
"grad_norm": 4.402308464050293,
|
| 523 |
+
"learning_rate": 6.222222222222222e-05,
|
| 524 |
+
"loss": 0.1246,
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 6.68,
|
| 529 |
+
"grad_norm": 4.047673225402832,
|
| 530 |
+
"learning_rate": 6.148148148148148e-05,
|
| 531 |
+
"loss": 0.1128,
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 6.78,
|
| 536 |
+
"grad_norm": 5.06707763671875,
|
| 537 |
+
"learning_rate": 6.074074074074074e-05,
|
| 538 |
+
"loss": 0.1315,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 6.88,
|
| 543 |
+
"grad_norm": 3.584838628768921,
|
| 544 |
+
"learning_rate": 6e-05,
|
| 545 |
+
"loss": 0.1052,
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 6.98,
|
| 550 |
+
"grad_norm": 4.084716320037842,
|
| 551 |
+
"learning_rate": 5.925925925925926e-05,
|
| 552 |
+
"loss": 0.1188,
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 6.99,
|
| 557 |
+
"eval_accuracy": 0.9046283309957924,
|
| 558 |
+
"eval_loss": 0.373639851808548,
|
| 559 |
+
"eval_runtime": 39.2003,
|
| 560 |
+
"eval_samples_per_second": 72.754,
|
| 561 |
+
"eval_steps_per_second": 1.148,
|
| 562 |
"step": 701
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 7.08,
|
| 566 |
+
"grad_norm": 5.199974536895752,
|
| 567 |
+
"learning_rate": 5.851851851851852e-05,
|
| 568 |
+
"loss": 0.1068,
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 7.18,
|
| 573 |
+
"grad_norm": 3.842390775680542,
|
| 574 |
+
"learning_rate": 5.7777777777777776e-05,
|
| 575 |
+
"loss": 0.1011,
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 7.28,
|
| 580 |
+
"grad_norm": 4.7214837074279785,
|
| 581 |
+
"learning_rate": 5.703703703703704e-05,
|
| 582 |
+
"loss": 0.1045,
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 7.38,
|
| 587 |
+
"grad_norm": 4.736299514770508,
|
| 588 |
+
"learning_rate": 5.62962962962963e-05,
|
| 589 |
+
"loss": 0.1071,
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 7.48,
|
| 594 |
+
"grad_norm": 5.003107070922852,
|
| 595 |
+
"learning_rate": 5.555555555555556e-05,
|
| 596 |
+
"loss": 0.0928,
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 7.58,
|
| 601 |
+
"grad_norm": 5.801096439361572,
|
| 602 |
+
"learning_rate": 5.4814814814814817e-05,
|
| 603 |
+
"loss": 0.118,
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 7.68,
|
| 608 |
+
"grad_norm": 5.874922752380371,
|
| 609 |
+
"learning_rate": 5.4074074074074075e-05,
|
| 610 |
+
"loss": 0.108,
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 7.78,
|
| 615 |
+
"grad_norm": 3.476767063140869,
|
| 616 |
+
"learning_rate": 5.333333333333333e-05,
|
| 617 |
+
"loss": 0.1034,
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 7.88,
|
| 622 |
+
"grad_norm": 4.573951244354248,
|
| 623 |
+
"learning_rate": 5.259259259259259e-05,
|
| 624 |
+
"loss": 0.121,
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 7.98,
|
| 629 |
+
"grad_norm": 4.446974754333496,
|
| 630 |
+
"learning_rate": 5.185185185185185e-05,
|
| 631 |
+
"loss": 0.1257,
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 8.0,
|
| 636 |
+
"eval_accuracy": 0.9102384291725105,
|
| 637 |
+
"eval_loss": 0.33806487917900085,
|
| 638 |
+
"eval_runtime": 39.6445,
|
| 639 |
+
"eval_samples_per_second": 71.939,
|
| 640 |
+
"eval_steps_per_second": 1.135,
|
| 641 |
"step": 802
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 8.08,
|
| 645 |
+
"grad_norm": 3.8721048831939697,
|
| 646 |
+
"learning_rate": 5.111111111111111e-05,
|
| 647 |
+
"loss": 0.1036,
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
"epoch": 8.18,
|
| 652 |
+
"grad_norm": 4.111603736877441,
|
| 653 |
+
"learning_rate": 5.0370370370370366e-05,
|
| 654 |
+
"loss": 0.1031,
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 8.28,
|
| 659 |
+
"grad_norm": 5.5718674659729,
|
| 660 |
+
"learning_rate": 4.962962962962963e-05,
|
| 661 |
+
"loss": 0.1057,
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 8.38,
|
| 666 |
+
"grad_norm": 5.053532123565674,
|
| 667 |
+
"learning_rate": 4.888888888888889e-05,
|
| 668 |
+
"loss": 0.1038,
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 8.48,
|
| 673 |
+
"grad_norm": 5.147861480712891,
|
| 674 |
+
"learning_rate": 4.814814814814815e-05,
|
| 675 |
+
"loss": 0.1089,
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 8.58,
|
| 680 |
+
"grad_norm": 3.2430307865142822,
|
| 681 |
+
"learning_rate": 4.740740740740741e-05,
|
| 682 |
+
"loss": 0.0912,
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 8.68,
|
| 687 |
+
"grad_norm": 4.054327011108398,
|
| 688 |
+
"learning_rate": 4.666666666666667e-05,
|
| 689 |
+
"loss": 0.0956,
|
| 690 |
"step": 870
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 8.78,
|
| 694 |
+
"grad_norm": 4.733587741851807,
|
| 695 |
+
"learning_rate": 4.592592592592593e-05,
|
| 696 |
+
"loss": 0.1178,
|
| 697 |
"step": 880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 8.88,
|
| 701 |
+
"grad_norm": 3.2026824951171875,
|
| 702 |
+
"learning_rate": 4.518518518518519e-05,
|
| 703 |
+
"loss": 0.0937,
|
| 704 |
"step": 890
|
| 705 |
},
|
| 706 |
{
|
| 707 |
"epoch": 8.98,
|
| 708 |
+
"grad_norm": 4.621606349945068,
|
| 709 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 710 |
+
"loss": 0.1017,
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 9.0,
|
| 715 |
+
"eval_accuracy": 0.9214586255259467,
|
| 716 |
+
"eval_loss": 0.2871844172477722,
|
| 717 |
+
"eval_runtime": 39.1384,
|
| 718 |
+
"eval_samples_per_second": 72.87,
|
| 719 |
+
"eval_steps_per_second": 1.15,
|
| 720 |
"step": 902
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 9.08,
|
| 724 |
+
"grad_norm": 4.2190141677856445,
|
| 725 |
+
"learning_rate": 4.3703703703703705e-05,
|
| 726 |
+
"loss": 0.0941,
|
| 727 |
"step": 910
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 9.18,
|
| 731 |
+
"grad_norm": 4.141015529632568,
|
| 732 |
+
"learning_rate": 4.296296296296296e-05,
|
| 733 |
+
"loss": 0.1028,
|
| 734 |
"step": 920
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 9.28,
|
| 738 |
+
"grad_norm": 2.8188929557800293,
|
| 739 |
+
"learning_rate": 4.222222222222222e-05,
|
| 740 |
+
"loss": 0.0947,
|
| 741 |
"step": 930
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 9.38,
|
| 745 |
+
"grad_norm": 4.885405540466309,
|
| 746 |
+
"learning_rate": 4.148148148148148e-05,
|
| 747 |
+
"loss": 0.0984,
|
| 748 |
"step": 940
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 9.48,
|
| 752 |
+
"grad_norm": 3.4018211364746094,
|
| 753 |
+
"learning_rate": 4.074074074074074e-05,
|
| 754 |
+
"loss": 0.1083,
|
| 755 |
"step": 950
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 9.58,
|
| 759 |
+
"grad_norm": 5.3954548835754395,
|
| 760 |
+
"learning_rate": 4e-05,
|
| 761 |
+
"loss": 0.0934,
|
| 762 |
"step": 960
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 9.68,
|
| 766 |
+
"grad_norm": 3.0029079914093018,
|
| 767 |
+
"learning_rate": 3.925925925925926e-05,
|
| 768 |
+
"loss": 0.1042,
|
| 769 |
"step": 970
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 9.78,
|
| 773 |
+
"grad_norm": 3.6331303119659424,
|
| 774 |
+
"learning_rate": 3.851851851851852e-05,
|
| 775 |
+
"loss": 0.1075,
|
| 776 |
"step": 980
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 9.88,
|
| 780 |
+
"grad_norm": 4.058956623077393,
|
| 781 |
+
"learning_rate": 3.777777777777778e-05,
|
| 782 |
+
"loss": 0.0988,
|
| 783 |
"step": 990
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 9.98,
|
| 787 |
+
"grad_norm": 4.427227020263672,
|
| 788 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 789 |
+
"loss": 0.0987,
|
| 790 |
"step": 1000
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"epoch": 10.0,
|
| 794 |
+
"eval_accuracy": 0.917601683029453,
|
| 795 |
+
"eval_loss": 0.30670177936553955,
|
| 796 |
+
"eval_runtime": 39.1776,
|
| 797 |
+
"eval_samples_per_second": 72.797,
|
| 798 |
+
"eval_steps_per_second": 1.149,
|
| 799 |
+
"step": 1002
|
| 800 |
},
|
| 801 |
{
|
| 802 |
+
"epoch": 10.07,
|
| 803 |
+
"grad_norm": 3.0564310550689697,
|
| 804 |
+
"learning_rate": 3.62962962962963e-05,
|
| 805 |
+
"loss": 0.099,
|
| 806 |
+
"step": 1010
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 10.17,
|
| 810 |
+
"grad_norm": 4.081722736358643,
|
| 811 |
+
"learning_rate": 3.555555555555556e-05,
|
| 812 |
+
"loss": 0.1069,
|
| 813 |
+
"step": 1020
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 10.27,
|
| 817 |
+
"grad_norm": 5.021610260009766,
|
| 818 |
+
"learning_rate": 3.481481481481482e-05,
|
| 819 |
+
"loss": 0.0963,
|
| 820 |
+
"step": 1030
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 10.37,
|
| 824 |
+
"grad_norm": 3.7664260864257812,
|
| 825 |
+
"learning_rate": 3.4074074074074077e-05,
|
| 826 |
+
"loss": 0.0857,
|
| 827 |
+
"step": 1040
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 10.47,
|
| 831 |
+
"grad_norm": 4.532868385314941,
|
| 832 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 833 |
+
"loss": 0.0945,
|
| 834 |
+
"step": 1050
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 10.57,
|
| 838 |
+
"grad_norm": 3.1503536701202393,
|
| 839 |
+
"learning_rate": 3.25925925925926e-05,
|
| 840 |
+
"loss": 0.1018,
|
| 841 |
+
"step": 1060
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 10.67,
|
| 845 |
+
"grad_norm": 4.589663982391357,
|
| 846 |
+
"learning_rate": 3.185185185185185e-05,
|
| 847 |
+
"loss": 0.0924,
|
| 848 |
+
"step": 1070
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 10.77,
|
| 852 |
+
"grad_norm": 2.68914794921875,
|
| 853 |
+
"learning_rate": 3.111111111111111e-05,
|
| 854 |
+
"loss": 0.0945,
|
| 855 |
+
"step": 1080
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 10.87,
|
| 859 |
+
"grad_norm": 3.398358106613159,
|
| 860 |
+
"learning_rate": 3.037037037037037e-05,
|
| 861 |
+
"loss": 0.0877,
|
| 862 |
+
"step": 1090
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 10.97,
|
| 866 |
+
"grad_norm": 4.018301010131836,
|
| 867 |
+
"learning_rate": 2.962962962962963e-05,
|
| 868 |
+
"loss": 0.0874,
|
| 869 |
+
"step": 1100
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 10.99,
|
| 873 |
+
"eval_accuracy": 0.9165497896213184,
|
| 874 |
+
"eval_loss": 0.29185158014297485,
|
| 875 |
+
"eval_runtime": 39.4428,
|
| 876 |
+
"eval_samples_per_second": 72.307,
|
| 877 |
+
"eval_steps_per_second": 1.141,
|
| 878 |
+
"step": 1102
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 11.07,
|
| 882 |
+
"grad_norm": 4.864216327667236,
|
| 883 |
+
"learning_rate": 2.8888888888888888e-05,
|
| 884 |
+
"loss": 0.0979,
|
| 885 |
+
"step": 1110
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 11.17,
|
| 889 |
+
"grad_norm": 5.510034084320068,
|
| 890 |
+
"learning_rate": 2.814814814814815e-05,
|
| 891 |
+
"loss": 0.0857,
|
| 892 |
+
"step": 1120
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 11.27,
|
| 896 |
+
"grad_norm": 6.076953887939453,
|
| 897 |
+
"learning_rate": 2.7407407407407408e-05,
|
| 898 |
+
"loss": 0.099,
|
| 899 |
+
"step": 1130
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 11.37,
|
| 903 |
+
"grad_norm": 3.9851274490356445,
|
| 904 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 905 |
+
"loss": 0.0967,
|
| 906 |
+
"step": 1140
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 11.47,
|
| 910 |
+
"grad_norm": 4.604448318481445,
|
| 911 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 912 |
+
"loss": 0.0975,
|
| 913 |
+
"step": 1150
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 11.57,
|
| 917 |
+
"grad_norm": 1.658200740814209,
|
| 918 |
+
"learning_rate": 2.5185185185185183e-05,
|
| 919 |
+
"loss": 0.0849,
|
| 920 |
+
"step": 1160
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 11.67,
|
| 924 |
+
"grad_norm": 5.351607322692871,
|
| 925 |
+
"learning_rate": 2.4444444444444445e-05,
|
| 926 |
+
"loss": 0.0979,
|
| 927 |
+
"step": 1170
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 11.77,
|
| 931 |
+
"grad_norm": 4.404117584228516,
|
| 932 |
+
"learning_rate": 2.3703703703703707e-05,
|
| 933 |
+
"loss": 0.0953,
|
| 934 |
+
"step": 1180
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 11.87,
|
| 938 |
+
"grad_norm": 5.584921836853027,
|
| 939 |
+
"learning_rate": 2.2962962962962965e-05,
|
| 940 |
+
"loss": 0.0803,
|
| 941 |
+
"step": 1190
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 11.97,
|
| 945 |
+
"grad_norm": 11.780359268188477,
|
| 946 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 947 |
+
"loss": 0.0901,
|
| 948 |
+
"step": 1200
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 12.0,
|
| 952 |
+
"eval_accuracy": 0.9228611500701263,
|
| 953 |
+
"eval_loss": 0.2942492961883545,
|
| 954 |
+
"eval_runtime": 39.3265,
|
| 955 |
+
"eval_samples_per_second": 72.521,
|
| 956 |
+
"eval_steps_per_second": 1.144,
|
| 957 |
+
"step": 1203
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 12.07,
|
| 961 |
+
"grad_norm": 3.900620460510254,
|
| 962 |
+
"learning_rate": 2.148148148148148e-05,
|
| 963 |
+
"loss": 0.09,
|
| 964 |
+
"step": 1210
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 12.17,
|
| 968 |
+
"grad_norm": 3.0629093647003174,
|
| 969 |
+
"learning_rate": 2.074074074074074e-05,
|
| 970 |
+
"loss": 0.0769,
|
| 971 |
+
"step": 1220
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 12.27,
|
| 975 |
+
"grad_norm": 3.7691752910614014,
|
| 976 |
+
"learning_rate": 2e-05,
|
| 977 |
+
"loss": 0.0925,
|
| 978 |
+
"step": 1230
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 12.37,
|
| 982 |
+
"grad_norm": 3.8815600872039795,
|
| 983 |
+
"learning_rate": 1.925925925925926e-05,
|
| 984 |
+
"loss": 0.0736,
|
| 985 |
+
"step": 1240
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 12.47,
|
| 989 |
+
"grad_norm": 3.6754331588745117,
|
| 990 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 991 |
+
"loss": 0.0826,
|
| 992 |
+
"step": 1250
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 12.57,
|
| 996 |
+
"grad_norm": 4.773692607879639,
|
| 997 |
+
"learning_rate": 1.777777777777778e-05,
|
| 998 |
+
"loss": 0.0821,
|
| 999 |
+
"step": 1260
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 12.67,
|
| 1003 |
+
"grad_norm": 4.411365985870361,
|
| 1004 |
+
"learning_rate": 1.7037037037037038e-05,
|
| 1005 |
+
"loss": 0.0754,
|
| 1006 |
+
"step": 1270
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 12.77,
|
| 1010 |
+
"grad_norm": 3.2157444953918457,
|
| 1011 |
+
"learning_rate": 1.62962962962963e-05,
|
| 1012 |
+
"loss": 0.0736,
|
| 1013 |
+
"step": 1280
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 12.87,
|
| 1017 |
+
"grad_norm": 3.8522512912750244,
|
| 1018 |
+
"learning_rate": 1.5555555555555555e-05,
|
| 1019 |
+
"loss": 0.0806,
|
| 1020 |
+
"step": 1290
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 12.97,
|
| 1024 |
+
"grad_norm": 3.8069818019866943,
|
| 1025 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 1026 |
+
"loss": 0.0831,
|
| 1027 |
+
"step": 1300
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 13.0,
|
| 1031 |
+
"eval_accuracy": 0.9232117812061711,
|
| 1032 |
+
"eval_loss": 0.29743239283561707,
|
| 1033 |
+
"eval_runtime": 39.4899,
|
| 1034 |
+
"eval_samples_per_second": 72.221,
|
| 1035 |
+
"eval_steps_per_second": 1.14,
|
| 1036 |
+
"step": 1303
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 13.07,
|
| 1040 |
+
"grad_norm": 2.675555467605591,
|
| 1041 |
+
"learning_rate": 1.4074074074074075e-05,
|
| 1042 |
+
"loss": 0.0687,
|
| 1043 |
+
"step": 1310
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 13.17,
|
| 1047 |
+
"grad_norm": 4.165126800537109,
|
| 1048 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 1049 |
+
"loss": 0.0694,
|
| 1050 |
+
"step": 1320
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 13.27,
|
| 1054 |
+
"grad_norm": 3.973789930343628,
|
| 1055 |
+
"learning_rate": 1.2592592592592592e-05,
|
| 1056 |
+
"loss": 0.0831,
|
| 1057 |
+
"step": 1330
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 13.37,
|
| 1061 |
+
"grad_norm": 3.488872528076172,
|
| 1062 |
+
"learning_rate": 1.1851851851851853e-05,
|
| 1063 |
+
"loss": 0.0761,
|
| 1064 |
+
"step": 1340
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 13.47,
|
| 1068 |
+
"grad_norm": 3.375296115875244,
|
| 1069 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 1070 |
+
"loss": 0.0607,
|
| 1071 |
+
"step": 1350
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 13.57,
|
| 1075 |
+
"grad_norm": 1.8006842136383057,
|
| 1076 |
+
"learning_rate": 1.037037037037037e-05,
|
| 1077 |
+
"loss": 0.0681,
|
| 1078 |
+
"step": 1360
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 13.67,
|
| 1082 |
+
"grad_norm": 3.174525260925293,
|
| 1083 |
+
"learning_rate": 9.62962962962963e-06,
|
| 1084 |
+
"loss": 0.0622,
|
| 1085 |
+
"step": 1370
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 13.77,
|
| 1089 |
+
"grad_norm": 3.6712663173675537,
|
| 1090 |
+
"learning_rate": 8.88888888888889e-06,
|
| 1091 |
+
"loss": 0.082,
|
| 1092 |
+
"step": 1380
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 13.87,
|
| 1096 |
+
"grad_norm": 2.8986759185791016,
|
| 1097 |
+
"learning_rate": 8.14814814814815e-06,
|
| 1098 |
+
"loss": 0.0879,
|
| 1099 |
+
"step": 1390
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 13.97,
|
| 1103 |
+
"grad_norm": 4.151699066162109,
|
| 1104 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 1105 |
+
"loss": 0.0838,
|
| 1106 |
+
"step": 1400
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 14.0,
|
| 1110 |
+
"eval_accuracy": 0.926367461430575,
|
| 1111 |
+
"eval_loss": 0.27871325612068176,
|
| 1112 |
+
"eval_runtime": 39.5001,
|
| 1113 |
+
"eval_samples_per_second": 72.202,
|
| 1114 |
+
"eval_steps_per_second": 1.139,
|
| 1115 |
+
"step": 1403
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 14.06,
|
| 1119 |
+
"grad_norm": 4.287971019744873,
|
| 1120 |
+
"learning_rate": 6.666666666666667e-06,
|
| 1121 |
+
"loss": 0.0731,
|
| 1122 |
+
"step": 1410
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 14.16,
|
| 1126 |
+
"grad_norm": 2.9228641986846924,
|
| 1127 |
+
"learning_rate": 5.925925925925927e-06,
|
| 1128 |
+
"loss": 0.0659,
|
| 1129 |
+
"step": 1420
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 14.26,
|
| 1133 |
+
"grad_norm": 1.4059169292449951,
|
| 1134 |
+
"learning_rate": 5.185185185185185e-06,
|
| 1135 |
+
"loss": 0.0654,
|
| 1136 |
+
"step": 1430
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 14.36,
|
| 1140 |
+
"grad_norm": 4.815260887145996,
|
| 1141 |
+
"learning_rate": 4.444444444444445e-06,
|
| 1142 |
+
"loss": 0.081,
|
| 1143 |
+
"step": 1440
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 14.46,
|
| 1147 |
+
"grad_norm": 2.322667121887207,
|
| 1148 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 1149 |
+
"loss": 0.0689,
|
| 1150 |
+
"step": 1450
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 14.56,
|
| 1154 |
+
"grad_norm": 4.125411033630371,
|
| 1155 |
+
"learning_rate": 2.9629629629629633e-06,
|
| 1156 |
+
"loss": 0.0716,
|
| 1157 |
+
"step": 1460
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 14.66,
|
| 1161 |
+
"grad_norm": 3.8639822006225586,
|
| 1162 |
+
"learning_rate": 2.2222222222222225e-06,
|
| 1163 |
+
"loss": 0.072,
|
| 1164 |
+
"step": 1470
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 14.76,
|
| 1168 |
+
"grad_norm": 1.4972394704818726,
|
| 1169 |
+
"learning_rate": 1.4814814814814817e-06,
|
| 1170 |
+
"loss": 0.0728,
|
| 1171 |
+
"step": 1480
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 14.86,
|
| 1175 |
+
"grad_norm": 1.8535915613174438,
|
| 1176 |
+
"learning_rate": 7.407407407407408e-07,
|
| 1177 |
+
"loss": 0.077,
|
| 1178 |
+
"step": 1490
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 14.96,
|
| 1182 |
+
"grad_norm": 4.621140003204346,
|
| 1183 |
+
"learning_rate": 0.0,
|
| 1184 |
+
"loss": 0.0603,
|
| 1185 |
+
"step": 1500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 14.96,
|
| 1189 |
+
"eval_accuracy": 0.926367461430575,
|
| 1190 |
+
"eval_loss": 0.2779853940010071,
|
| 1191 |
+
"eval_runtime": 39.6501,
|
| 1192 |
+
"eval_samples_per_second": 71.929,
|
| 1193 |
+
"eval_steps_per_second": 1.135,
|
| 1194 |
+
"step": 1500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 14.96,
|
| 1198 |
+
"step": 1500,
|
| 1199 |
+
"total_flos": 9.546299981955072e+18,
|
| 1200 |
+
"train_loss": 0.11206140112876892,
|
| 1201 |
+
"train_runtime": 7336.5108,
|
| 1202 |
+
"train_samples_per_second": 52.472,
|
| 1203 |
"train_steps_per_second": 0.204
|
| 1204 |
}
|
| 1205 |
],
|
| 1206 |
"logging_steps": 10,
|
| 1207 |
+
"max_steps": 1500,
|
| 1208 |
"num_input_tokens_seen": 0,
|
| 1209 |
+
"num_train_epochs": 15,
|
| 1210 |
"save_steps": 500,
|
| 1211 |
+
"total_flos": 9.546299981955072e+18,
|
| 1212 |
"train_batch_size": 64,
|
| 1213 |
"trial_name": null,
|
| 1214 |
"trial_params": null
|