Commit
•
95ec253
1
Parent(s):
d6409de
End of training
Browse files- README.md +3 -3
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Apr30_00-59-58_786022f41cde/events.out.tfevents.1714445352.786022f41cde.42.6 +3 -0
- train_results.json +6 -6
- trainer_state.json +982 -265
README.md
CHANGED
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the image_folder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss:
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.7057676232933965
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the image_folder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.9803
|
| 36 |
+
- Accuracy: 0.7058
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 60.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second": 44.
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.98,
|
| 3 |
+
"eval_accuracy": 0.7057676232933965,
|
| 4 |
+
"eval_loss": 0.9803113341331482,
|
| 5 |
+
"eval_runtime": 60.2511,
|
| 6 |
+
"eval_samples_per_second": 119.135,
|
| 7 |
+
"eval_steps_per_second": 3.734,
|
| 8 |
+
"total_flos": 2.2200667552042852e+19,
|
| 9 |
+
"train_loss": 0.27600424638284105,
|
| 10 |
+
"train_runtime": 6453.2956,
|
| 11 |
+
"train_samples_per_second": 44.487,
|
| 12 |
+
"train_steps_per_second": 0.347
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 60.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.98,
|
| 3 |
+
"eval_accuracy": 0.7057676232933965,
|
| 4 |
+
"eval_loss": 0.9803113341331482,
|
| 5 |
+
"eval_runtime": 60.2511,
|
| 6 |
+
"eval_samples_per_second": 119.135,
|
| 7 |
+
"eval_steps_per_second": 3.734
|
| 8 |
}
|
runs/Apr30_00-59-58_786022f41cde/events.out.tfevents.1714445352.786022f41cde.42.6
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aaf8dc49760c3ee0adee024a5228dc39f7dace6ad559e1699aef2e9652b8975c
|
| 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": 44.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.98,
|
| 3 |
+
"total_flos": 2.2200667552042852e+19,
|
| 4 |
+
"train_loss": 0.27600424638284105,
|
| 5 |
+
"train_runtime": 6453.2956,
|
| 6 |
+
"train_samples_per_second": 44.487,
|
| 7 |
+
"train_steps_per_second": 0.347
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,746 +1,1463 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned/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": 0.04,
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss":
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.09,
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss":
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.13,
|
| 25 |
-
"learning_rate":
|
| 26 |
-
"loss":
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.18,
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss":
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.22,
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss": 0.
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.27,
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.31,
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.36,
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"loss": 0.
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.4,
|
| 61 |
-
"learning_rate":
|
| 62 |
-
"loss": 0.
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.45,
|
| 67 |
-
"learning_rate":
|
| 68 |
-
"loss": 0.
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.49,
|
| 73 |
-
"learning_rate":
|
| 74 |
-
"loss": 0.
|
| 75 |
"step": 110
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.53,
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 120
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.58,
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 130
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.62,
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 140
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.67,
|
| 97 |
-
"learning_rate":
|
| 98 |
-
"loss": 0.
|
| 99 |
"step": 150
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.71,
|
| 103 |
-
"learning_rate":
|
| 104 |
-
"loss": 0.
|
| 105 |
"step": 160
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 0.76,
|
| 109 |
-
"learning_rate":
|
| 110 |
-
"loss": 0.
|
| 111 |
"step": 170
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.8,
|
| 115 |
-
"learning_rate": 4.
|
| 116 |
-
"loss": 0.
|
| 117 |
"step": 180
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 0.85,
|
| 121 |
-
"learning_rate": 4.
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 190
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.89,
|
| 127 |
-
"learning_rate": 4.
|
| 128 |
-
"loss": 0.
|
| 129 |
"step": 200
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.94,
|
| 133 |
-
"learning_rate": 4.
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 210
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.98,
|
| 139 |
-
"learning_rate": 4.
|
| 140 |
-
"loss": 0.
|
| 141 |
"step": 220
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 1.0,
|
| 145 |
-
"eval_accuracy": 0.
|
| 146 |
-
"eval_loss": 0.
|
| 147 |
-
"eval_runtime": 60.
|
| 148 |
-
"eval_samples_per_second": 119.
|
| 149 |
-
"eval_steps_per_second": 3.
|
| 150 |
"step": 224
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 1.02,
|
| 154 |
-
"learning_rate": 4.
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 230
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 1.07,
|
| 160 |
-
"learning_rate": 4.
|
| 161 |
-
"loss": 0.
|
| 162 |
"step": 240
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 1.11,
|
| 166 |
-
"learning_rate": 4.
|
| 167 |
-
"loss": 0.
|
| 168 |
"step": 250
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 1.16,
|
| 172 |
-
"learning_rate": 4.
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 260
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 1.2,
|
| 178 |
-
"learning_rate": 4.
|
| 179 |
-
"loss": 0.
|
| 180 |
"step": 270
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 1.25,
|
| 184 |
-
"learning_rate": 4.
|
| 185 |
-
"loss": 0.
|
| 186 |
"step": 280
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 1.29,
|
| 190 |
-
"learning_rate": 4.
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 290
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 1.34,
|
| 196 |
-
"learning_rate": 4.
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 300
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 1.38,
|
| 202 |
-
"learning_rate": 4.
|
| 203 |
-
"loss": 0.
|
| 204 |
"step": 310
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 1.43,
|
| 208 |
-
"learning_rate":
|
| 209 |
-
"loss": 0.
|
| 210 |
"step": 320
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 1.47,
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 330
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 1.51,
|
| 220 |
-
"learning_rate":
|
| 221 |
-
"loss": 0.
|
| 222 |
"step": 340
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 1.56,
|
| 226 |
-
"learning_rate":
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 350
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 1.6,
|
| 232 |
-
"learning_rate":
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 360
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 1.65,
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 370
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 1.69,
|
| 244 |
-
"learning_rate":
|
| 245 |
-
"loss": 0.
|
| 246 |
"step": 380
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 1.74,
|
| 250 |
-
"learning_rate":
|
| 251 |
-
"loss": 0.
|
| 252 |
"step": 390
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 1.78,
|
| 256 |
-
"learning_rate":
|
| 257 |
-
"loss": 0.
|
| 258 |
"step": 400
|
| 259 |
},
|
| 260 |
{
|
| 261 |
"epoch": 1.83,
|
| 262 |
-
"learning_rate":
|
| 263 |
-
"loss": 0.
|
| 264 |
"step": 410
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 1.87,
|
| 268 |
-
"learning_rate":
|
| 269 |
-
"loss": 0.
|
| 270 |
"step": 420
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 1.92,
|
| 274 |
-
"learning_rate":
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 430
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 1.96,
|
| 280 |
-
"learning_rate":
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 440
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 2.0,
|
| 286 |
-
"eval_accuracy": 0.
|
| 287 |
-
"eval_loss": 0.
|
| 288 |
-
"eval_runtime":
|
| 289 |
-
"eval_samples_per_second":
|
| 290 |
-
"eval_steps_per_second": 3.
|
| 291 |
"step": 449
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 2.0,
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 450
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 2.05,
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 460
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 2.09,
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 470
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 2.14,
|
| 313 |
-
"learning_rate":
|
| 314 |
-
"loss": 0.
|
| 315 |
"step": 480
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 2.18,
|
| 319 |
-
"learning_rate":
|
| 320 |
-
"loss": 0.
|
| 321 |
"step": 490
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 2.23,
|
| 325 |
-
"learning_rate":
|
| 326 |
-
"loss": 0.
|
| 327 |
"step": 500
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 2.27,
|
| 331 |
-
"learning_rate":
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 510
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 2.32,
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 520
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 2.36,
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 530
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 2.41,
|
| 349 |
-
"learning_rate":
|
| 350 |
-
"loss": 0.
|
| 351 |
"step": 540
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 2.45,
|
| 355 |
-
"learning_rate":
|
| 356 |
-
"loss": 0.
|
| 357 |
"step": 550
|
| 358 |
},
|
| 359 |
{
|
| 360 |
"epoch": 2.49,
|
| 361 |
-
"learning_rate":
|
| 362 |
-
"loss": 0.
|
| 363 |
"step": 560
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 2.54,
|
| 367 |
-
"learning_rate":
|
| 368 |
-
"loss": 0.
|
| 369 |
"step": 570
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 2.58,
|
| 373 |
-
"learning_rate":
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 580
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 2.63,
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 590
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 2.67,
|
| 385 |
-
"learning_rate":
|
| 386 |
-
"loss": 0.
|
| 387 |
"step": 600
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 2.72,
|
| 391 |
-
"learning_rate":
|
| 392 |
-
"loss": 0.
|
| 393 |
"step": 610
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 2.76,
|
| 397 |
-
"learning_rate":
|
| 398 |
-
"loss": 0.
|
| 399 |
"step": 620
|
| 400 |
},
|
| 401 |
{
|
| 402 |
"epoch": 2.81,
|
| 403 |
-
"learning_rate":
|
| 404 |
-
"loss": 0.
|
| 405 |
"step": 630
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 2.85,
|
| 409 |
-
"learning_rate":
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 640
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 2.9,
|
| 415 |
-
"learning_rate":
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 650
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 2.94,
|
| 421 |
-
"learning_rate":
|
| 422 |
-
"loss": 0.
|
| 423 |
"step": 660
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 2.98,
|
| 427 |
-
"learning_rate":
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 670
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 3.0,
|
| 433 |
-
"eval_accuracy": 0.
|
| 434 |
-
"eval_loss": 0.
|
| 435 |
-
"eval_runtime":
|
| 436 |
-
"eval_samples_per_second":
|
| 437 |
-
"eval_steps_per_second": 3.
|
| 438 |
"step": 673
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 3.03,
|
| 442 |
-
"learning_rate":
|
| 443 |
-
"loss": 0.
|
| 444 |
"step": 680
|
| 445 |
},
|
| 446 |
{
|
| 447 |
"epoch": 3.07,
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss": 0.
|
| 450 |
"step": 690
|
| 451 |
},
|
| 452 |
{
|
| 453 |
"epoch": 3.12,
|
| 454 |
-
"learning_rate":
|
| 455 |
-
"loss": 0.
|
| 456 |
"step": 700
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 3.16,
|
| 460 |
-
"learning_rate":
|
| 461 |
-
"loss": 0.
|
| 462 |
"step": 710
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 3.21,
|
| 466 |
-
"learning_rate":
|
| 467 |
-
"loss": 0.
|
| 468 |
"step": 720
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 3.25,
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 730
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 3.3,
|
| 478 |
-
"learning_rate":
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 740
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 3.34,
|
| 484 |
-
"learning_rate":
|
| 485 |
-
"loss": 0.
|
| 486 |
"step": 750
|
| 487 |
},
|
| 488 |
{
|
| 489 |
"epoch": 3.39,
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss": 0.
|
| 492 |
"step": 760
|
| 493 |
},
|
| 494 |
{
|
| 495 |
"epoch": 3.43,
|
| 496 |
-
"learning_rate":
|
| 497 |
-
"loss": 0.
|
| 498 |
"step": 770
|
| 499 |
},
|
| 500 |
{
|
| 501 |
"epoch": 3.47,
|
| 502 |
-
"learning_rate":
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 780
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 3.52,
|
| 508 |
-
"learning_rate":
|
| 509 |
-
"loss": 0.
|
| 510 |
"step": 790
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 3.56,
|
| 514 |
-
"learning_rate":
|
| 515 |
-
"loss": 0.
|
| 516 |
"step": 800
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 3.61,
|
| 520 |
-
"learning_rate":
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 810
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 3.65,
|
| 526 |
-
"learning_rate":
|
| 527 |
-
"loss": 0.
|
| 528 |
"step": 820
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 3.7,
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 830
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 3.74,
|
| 538 |
-
"learning_rate":
|
| 539 |
-
"loss": 0.
|
| 540 |
"step": 840
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 3.79,
|
| 544 |
-
"learning_rate":
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 850
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 3.83,
|
| 550 |
-
"learning_rate":
|
| 551 |
-
"loss": 0.
|
| 552 |
"step": 860
|
| 553 |
},
|
| 554 |
{
|
| 555 |
"epoch": 3.88,
|
| 556 |
-
"learning_rate":
|
| 557 |
-
"loss": 0.
|
| 558 |
"step": 870
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 3.92,
|
| 562 |
-
"learning_rate":
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 880
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 3.96,
|
| 568 |
-
"learning_rate":
|
| 569 |
-
"loss": 0.
|
| 570 |
"step": 890
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 4.0,
|
| 574 |
-
"eval_accuracy": 0.
|
| 575 |
-
"eval_loss": 0.
|
| 576 |
-
"eval_runtime":
|
| 577 |
-
"eval_samples_per_second":
|
| 578 |
-
"eval_steps_per_second": 3.
|
| 579 |
"step": 898
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 4.01,
|
| 583 |
-
"learning_rate":
|
| 584 |
-
"loss": 0.
|
| 585 |
"step": 900
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 4.05,
|
| 589 |
-
"learning_rate":
|
| 590 |
-
"loss": 0.
|
| 591 |
"step": 910
|
| 592 |
},
|
| 593 |
{
|
| 594 |
"epoch": 4.1,
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 920
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 4.14,
|
| 601 |
-
"learning_rate":
|
| 602 |
-
"loss": 0.
|
| 603 |
"step": 930
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 4.19,
|
| 607 |
-
"learning_rate":
|
| 608 |
-
"loss": 0.
|
| 609 |
"step": 940
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 4.23,
|
| 613 |
-
"learning_rate":
|
| 614 |
-
"loss": 0.
|
| 615 |
"step": 950
|
| 616 |
},
|
| 617 |
{
|
| 618 |
"epoch": 4.28,
|
| 619 |
-
"learning_rate":
|
| 620 |
-
"loss": 0.
|
| 621 |
"step": 960
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 4.32,
|
| 625 |
-
"learning_rate":
|
| 626 |
-
"loss": 0.
|
| 627 |
"step": 970
|
| 628 |
},
|
| 629 |
{
|
| 630 |
"epoch": 4.37,
|
| 631 |
-
"learning_rate":
|
| 632 |
-
"loss": 0.
|
| 633 |
"step": 980
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 4.41,
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 990
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 4.45,
|
| 643 |
-
"learning_rate":
|
| 644 |
-
"loss": 0.
|
| 645 |
"step": 1000
|
| 646 |
},
|
| 647 |
{
|
| 648 |
"epoch": 4.5,
|
| 649 |
-
"learning_rate":
|
| 650 |
-
"loss": 0.
|
| 651 |
"step": 1010
|
| 652 |
},
|
| 653 |
{
|
| 654 |
"epoch": 4.54,
|
| 655 |
-
"learning_rate":
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 1020
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 4.59,
|
| 661 |
-
"learning_rate":
|
| 662 |
-
"loss": 0.
|
| 663 |
"step": 1030
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 4.63,
|
| 667 |
-
"learning_rate":
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 1040
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 4.68,
|
| 673 |
-
"learning_rate":
|
| 674 |
-
"loss": 0.
|
| 675 |
"step": 1050
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 4.72,
|
| 679 |
-
"learning_rate": 2.
|
| 680 |
-
"loss": 0.
|
| 681 |
"step": 1060
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 4.77,
|
| 685 |
-
"learning_rate": 2.
|
| 686 |
-
"loss": 0.
|
| 687 |
"step": 1070
|
| 688 |
},
|
| 689 |
{
|
| 690 |
"epoch": 4.81,
|
| 691 |
-
"learning_rate":
|
| 692 |
-
"loss": 0.
|
| 693 |
"step": 1080
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 4.86,
|
| 697 |
-
"learning_rate":
|
| 698 |
-
"loss": 0.
|
| 699 |
"step": 1090
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 4.9,
|
| 703 |
-
"learning_rate":
|
| 704 |
-
"loss": 0.
|
| 705 |
"step": 1100
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 4.94,
|
| 709 |
-
"learning_rate":
|
| 710 |
-
"loss": 0.
|
| 711 |
"step": 1110
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 4.99,
|
| 715 |
-
"learning_rate":
|
| 716 |
-
"loss": 0.
|
| 717 |
"step": 1120
|
| 718 |
},
|
| 719 |
{
|
| 720 |
-
"epoch":
|
| 721 |
-
"eval_accuracy": 0.
|
| 722 |
-
"eval_loss": 0.
|
| 723 |
-
"eval_runtime": 60.
|
| 724 |
-
"eval_samples_per_second": 118.
|
| 725 |
-
"eval_steps_per_second": 3.
|
| 726 |
-
"step":
|
| 727 |
},
|
| 728 |
{
|
| 729 |
-
"epoch":
|
| 730 |
-
"
|
| 731 |
-
"
|
| 732 |
-
"
|
| 733 |
-
|
| 734 |
-
|
| 735 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 736 |
}
|
| 737 |
],
|
| 738 |
"logging_steps": 10,
|
| 739 |
-
"max_steps":
|
| 740 |
"num_input_tokens_seen": 0,
|
| 741 |
-
"num_train_epochs":
|
| 742 |
"save_steps": 500,
|
| 743 |
-
"total_flos":
|
| 744 |
"train_batch_size": 32,
|
| 745 |
"trial_name": null,
|
| 746 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.7057676232933965,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned/checkpoint-1122",
|
| 4 |
+
"epoch": 9.977728285077951,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 2240,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 2.2321428571428573e-06,
|
| 14 |
+
"loss": 0.5702,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.09,
|
| 19 |
+
"learning_rate": 4.464285714285715e-06,
|
| 20 |
+
"loss": 0.5531,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.13,
|
| 25 |
+
"learning_rate": 6.696428571428572e-06,
|
| 26 |
+
"loss": 0.5328,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.18,
|
| 31 |
+
"learning_rate": 8.92857142857143e-06,
|
| 32 |
+
"loss": 0.5412,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.22,
|
| 37 |
+
"learning_rate": 1.1160714285714287e-05,
|
| 38 |
+
"loss": 0.5131,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.27,
|
| 43 |
+
"learning_rate": 1.3392857142857144e-05,
|
| 44 |
+
"loss": 0.5021,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.31,
|
| 49 |
+
"learning_rate": 1.5625e-05,
|
| 50 |
+
"loss": 0.4657,
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.36,
|
| 55 |
+
"learning_rate": 1.785714285714286e-05,
|
| 56 |
+
"loss": 0.4831,
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.4,
|
| 61 |
+
"learning_rate": 2.0089285714285717e-05,
|
| 62 |
+
"loss": 0.4942,
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.45,
|
| 67 |
+
"learning_rate": 2.2321428571428575e-05,
|
| 68 |
+
"loss": 0.531,
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.49,
|
| 73 |
+
"learning_rate": 2.455357142857143e-05,
|
| 74 |
+
"loss": 0.4907,
|
| 75 |
"step": 110
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.53,
|
| 79 |
+
"learning_rate": 2.6785714285714288e-05,
|
| 80 |
+
"loss": 0.4482,
|
| 81 |
"step": 120
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.58,
|
| 85 |
+
"learning_rate": 2.9017857142857146e-05,
|
| 86 |
+
"loss": 0.5041,
|
| 87 |
"step": 130
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.62,
|
| 91 |
+
"learning_rate": 3.125e-05,
|
| 92 |
+
"loss": 0.4777,
|
| 93 |
"step": 140
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.67,
|
| 97 |
+
"learning_rate": 3.348214285714286e-05,
|
| 98 |
+
"loss": 0.4692,
|
| 99 |
"step": 150
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.71,
|
| 103 |
+
"learning_rate": 3.571428571428572e-05,
|
| 104 |
+
"loss": 0.4646,
|
| 105 |
"step": 160
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 0.76,
|
| 109 |
+
"learning_rate": 3.794642857142857e-05,
|
| 110 |
+
"loss": 0.4503,
|
| 111 |
"step": 170
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.8,
|
| 115 |
+
"learning_rate": 4.017857142857143e-05,
|
| 116 |
+
"loss": 0.5269,
|
| 117 |
"step": 180
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 0.85,
|
| 121 |
+
"learning_rate": 4.2410714285714285e-05,
|
| 122 |
+
"loss": 0.4849,
|
| 123 |
"step": 190
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.89,
|
| 127 |
+
"learning_rate": 4.464285714285715e-05,
|
| 128 |
+
"loss": 0.4956,
|
| 129 |
"step": 200
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.94,
|
| 133 |
+
"learning_rate": 4.6875e-05,
|
| 134 |
+
"loss": 0.4993,
|
| 135 |
"step": 210
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.98,
|
| 139 |
+
"learning_rate": 4.910714285714286e-05,
|
| 140 |
+
"loss": 0.4887,
|
| 141 |
"step": 220
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 1.0,
|
| 145 |
+
"eval_accuracy": 0.6776260796879353,
|
| 146 |
+
"eval_loss": 0.9212561845779419,
|
| 147 |
+
"eval_runtime": 60.2714,
|
| 148 |
+
"eval_samples_per_second": 119.095,
|
| 149 |
+
"eval_steps_per_second": 3.733,
|
| 150 |
"step": 224
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 1.02,
|
| 154 |
+
"learning_rate": 4.985119047619048e-05,
|
| 155 |
+
"loss": 0.5157,
|
| 156 |
"step": 230
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 1.07,
|
| 160 |
+
"learning_rate": 4.960317460317461e-05,
|
| 161 |
+
"loss": 0.5364,
|
| 162 |
"step": 240
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 1.11,
|
| 166 |
+
"learning_rate": 4.9355158730158735e-05,
|
| 167 |
+
"loss": 0.5125,
|
| 168 |
"step": 250
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 1.16,
|
| 172 |
+
"learning_rate": 4.910714285714286e-05,
|
| 173 |
+
"loss": 0.5028,
|
| 174 |
"step": 260
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 1.2,
|
| 178 |
+
"learning_rate": 4.8859126984126984e-05,
|
| 179 |
+
"loss": 0.4801,
|
| 180 |
"step": 270
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 1.25,
|
| 184 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 185 |
+
"loss": 0.5442,
|
| 186 |
"step": 280
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 1.29,
|
| 190 |
+
"learning_rate": 4.836309523809524e-05,
|
| 191 |
+
"loss": 0.4718,
|
| 192 |
"step": 290
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 1.34,
|
| 196 |
+
"learning_rate": 4.811507936507937e-05,
|
| 197 |
+
"loss": 0.5095,
|
| 198 |
"step": 300
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 1.38,
|
| 202 |
+
"learning_rate": 4.7867063492063496e-05,
|
| 203 |
+
"loss": 0.5151,
|
| 204 |
"step": 310
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 1.43,
|
| 208 |
+
"learning_rate": 4.761904761904762e-05,
|
| 209 |
+
"loss": 0.4909,
|
| 210 |
"step": 320
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 1.47,
|
| 214 |
+
"learning_rate": 4.7371031746031745e-05,
|
| 215 |
+
"loss": 0.4797,
|
| 216 |
"step": 330
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 1.51,
|
| 220 |
+
"learning_rate": 4.7123015873015876e-05,
|
| 221 |
+
"loss": 0.4604,
|
| 222 |
"step": 340
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 1.56,
|
| 226 |
+
"learning_rate": 4.6875e-05,
|
| 227 |
+
"loss": 0.5055,
|
| 228 |
"step": 350
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 1.6,
|
| 232 |
+
"learning_rate": 4.662698412698413e-05,
|
| 233 |
+
"loss": 0.5105,
|
| 234 |
"step": 360
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 1.65,
|
| 238 |
+
"learning_rate": 4.637896825396826e-05,
|
| 239 |
+
"loss": 0.4788,
|
| 240 |
"step": 370
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 1.69,
|
| 244 |
+
"learning_rate": 4.613095238095239e-05,
|
| 245 |
+
"loss": 0.4973,
|
| 246 |
"step": 380
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 1.74,
|
| 250 |
+
"learning_rate": 4.5882936507936506e-05,
|
| 251 |
+
"loss": 0.5572,
|
| 252 |
"step": 390
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 1.78,
|
| 256 |
+
"learning_rate": 4.563492063492064e-05,
|
| 257 |
+
"loss": 0.5182,
|
| 258 |
"step": 400
|
| 259 |
},
|
| 260 |
{
|
| 261 |
"epoch": 1.83,
|
| 262 |
+
"learning_rate": 4.538690476190476e-05,
|
| 263 |
+
"loss": 0.4698,
|
| 264 |
"step": 410
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 1.87,
|
| 268 |
+
"learning_rate": 4.5138888888888894e-05,
|
| 269 |
+
"loss": 0.4881,
|
| 270 |
"step": 420
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 1.92,
|
| 274 |
+
"learning_rate": 4.489087301587302e-05,
|
| 275 |
+
"loss": 0.4835,
|
| 276 |
"step": 430
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 1.96,
|
| 280 |
+
"learning_rate": 4.464285714285715e-05,
|
| 281 |
+
"loss": 0.4969,
|
| 282 |
"step": 440
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 2.0,
|
| 286 |
+
"eval_accuracy": 0.6926720534967957,
|
| 287 |
+
"eval_loss": 0.9037817716598511,
|
| 288 |
+
"eval_runtime": 60.4491,
|
| 289 |
+
"eval_samples_per_second": 118.744,
|
| 290 |
+
"eval_steps_per_second": 3.722,
|
| 291 |
"step": 449
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 2.0,
|
| 295 |
+
"learning_rate": 4.439484126984127e-05,
|
| 296 |
+
"loss": 0.4971,
|
| 297 |
"step": 450
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 2.05,
|
| 301 |
+
"learning_rate": 4.41468253968254e-05,
|
| 302 |
+
"loss": 0.4229,
|
| 303 |
"step": 460
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 2.09,
|
| 307 |
+
"learning_rate": 4.3898809523809523e-05,
|
| 308 |
+
"loss": 0.3694,
|
| 309 |
"step": 470
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 2.14,
|
| 313 |
+
"learning_rate": 4.3650793650793655e-05,
|
| 314 |
+
"loss": 0.4295,
|
| 315 |
"step": 480
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 2.18,
|
| 319 |
+
"learning_rate": 4.340277777777778e-05,
|
| 320 |
+
"loss": 0.3998,
|
| 321 |
"step": 490
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 2.23,
|
| 325 |
+
"learning_rate": 4.315476190476191e-05,
|
| 326 |
+
"loss": 0.4112,
|
| 327 |
"step": 500
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 2.27,
|
| 331 |
+
"learning_rate": 4.290674603174603e-05,
|
| 332 |
+
"loss": 0.3728,
|
| 333 |
"step": 510
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 2.32,
|
| 337 |
+
"learning_rate": 4.265873015873016e-05,
|
| 338 |
+
"loss": 0.4243,
|
| 339 |
"step": 520
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 2.36,
|
| 343 |
+
"learning_rate": 4.2410714285714285e-05,
|
| 344 |
+
"loss": 0.4209,
|
| 345 |
"step": 530
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 2.41,
|
| 349 |
+
"learning_rate": 4.2162698412698416e-05,
|
| 350 |
+
"loss": 0.3886,
|
| 351 |
"step": 540
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 2.45,
|
| 355 |
+
"learning_rate": 4.191468253968254e-05,
|
| 356 |
+
"loss": 0.3707,
|
| 357 |
"step": 550
|
| 358 |
},
|
| 359 |
{
|
| 360 |
"epoch": 2.49,
|
| 361 |
+
"learning_rate": 4.166666666666667e-05,
|
| 362 |
+
"loss": 0.4203,
|
| 363 |
"step": 560
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 2.54,
|
| 367 |
+
"learning_rate": 4.14186507936508e-05,
|
| 368 |
+
"loss": 0.4213,
|
| 369 |
"step": 570
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 2.58,
|
| 373 |
+
"learning_rate": 4.117063492063492e-05,
|
| 374 |
+
"loss": 0.3677,
|
| 375 |
"step": 580
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 2.63,
|
| 379 |
+
"learning_rate": 4.0922619047619046e-05,
|
| 380 |
+
"loss": 0.4198,
|
| 381 |
"step": 590
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 2.67,
|
| 385 |
+
"learning_rate": 4.067460317460318e-05,
|
| 386 |
+
"loss": 0.4307,
|
| 387 |
"step": 600
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 2.72,
|
| 391 |
+
"learning_rate": 4.04265873015873e-05,
|
| 392 |
+
"loss": 0.4073,
|
| 393 |
"step": 610
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 2.76,
|
| 397 |
+
"learning_rate": 4.017857142857143e-05,
|
| 398 |
+
"loss": 0.3957,
|
| 399 |
"step": 620
|
| 400 |
},
|
| 401 |
{
|
| 402 |
"epoch": 2.81,
|
| 403 |
+
"learning_rate": 3.993055555555556e-05,
|
| 404 |
+
"loss": 0.479,
|
| 405 |
"step": 630
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 2.85,
|
| 409 |
+
"learning_rate": 3.968253968253968e-05,
|
| 410 |
+
"loss": 0.4069,
|
| 411 |
"step": 640
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 2.9,
|
| 415 |
+
"learning_rate": 3.943452380952381e-05,
|
| 416 |
+
"loss": 0.417,
|
| 417 |
"step": 650
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 2.94,
|
| 421 |
+
"learning_rate": 3.918650793650794e-05,
|
| 422 |
+
"loss": 0.4129,
|
| 423 |
"step": 660
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 2.98,
|
| 427 |
+
"learning_rate": 3.893849206349206e-05,
|
| 428 |
+
"loss": 0.4095,
|
| 429 |
"step": 670
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 3.0,
|
| 433 |
+
"eval_accuracy": 0.6976873780997492,
|
| 434 |
+
"eval_loss": 0.9076758027076721,
|
| 435 |
+
"eval_runtime": 60.0956,
|
| 436 |
+
"eval_samples_per_second": 119.443,
|
| 437 |
+
"eval_steps_per_second": 3.744,
|
| 438 |
"step": 673
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 3.03,
|
| 442 |
+
"learning_rate": 3.8690476190476195e-05,
|
| 443 |
+
"loss": 0.3228,
|
| 444 |
"step": 680
|
| 445 |
},
|
| 446 |
{
|
| 447 |
"epoch": 3.07,
|
| 448 |
+
"learning_rate": 3.844246031746032e-05,
|
| 449 |
+
"loss": 0.3362,
|
| 450 |
"step": 690
|
| 451 |
},
|
| 452 |
{
|
| 453 |
"epoch": 3.12,
|
| 454 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 455 |
+
"loss": 0.3026,
|
| 456 |
"step": 700
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 3.16,
|
| 460 |
+
"learning_rate": 3.794642857142857e-05,
|
| 461 |
+
"loss": 0.3237,
|
| 462 |
"step": 710
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 3.21,
|
| 466 |
+
"learning_rate": 3.76984126984127e-05,
|
| 467 |
+
"loss": 0.3192,
|
| 468 |
"step": 720
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 3.25,
|
| 472 |
+
"learning_rate": 3.7450396825396824e-05,
|
| 473 |
+
"loss": 0.3461,
|
| 474 |
"step": 730
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 3.3,
|
| 478 |
+
"learning_rate": 3.7202380952380956e-05,
|
| 479 |
+
"loss": 0.3568,
|
| 480 |
"step": 740
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 3.34,
|
| 484 |
+
"learning_rate": 3.695436507936508e-05,
|
| 485 |
+
"loss": 0.3143,
|
| 486 |
"step": 750
|
| 487 |
},
|
| 488 |
{
|
| 489 |
"epoch": 3.39,
|
| 490 |
+
"learning_rate": 3.6706349206349205e-05,
|
| 491 |
+
"loss": 0.3381,
|
| 492 |
"step": 760
|
| 493 |
},
|
| 494 |
{
|
| 495 |
"epoch": 3.43,
|
| 496 |
+
"learning_rate": 3.6458333333333336e-05,
|
| 497 |
+
"loss": 0.3297,
|
| 498 |
"step": 770
|
| 499 |
},
|
| 500 |
{
|
| 501 |
"epoch": 3.47,
|
| 502 |
+
"learning_rate": 3.621031746031746e-05,
|
| 503 |
+
"loss": 0.3333,
|
| 504 |
"step": 780
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 3.52,
|
| 508 |
+
"learning_rate": 3.5962301587301586e-05,
|
| 509 |
+
"loss": 0.3232,
|
| 510 |
"step": 790
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 3.56,
|
| 514 |
+
"learning_rate": 3.571428571428572e-05,
|
| 515 |
+
"loss": 0.3449,
|
| 516 |
"step": 800
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 3.61,
|
| 520 |
+
"learning_rate": 3.546626984126984e-05,
|
| 521 |
+
"loss": 0.3563,
|
| 522 |
"step": 810
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 3.65,
|
| 526 |
+
"learning_rate": 3.521825396825397e-05,
|
| 527 |
+
"loss": 0.3326,
|
| 528 |
"step": 820
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 3.7,
|
| 532 |
+
"learning_rate": 3.49702380952381e-05,
|
| 533 |
+
"loss": 0.333,
|
| 534 |
"step": 830
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 3.74,
|
| 538 |
+
"learning_rate": 3.472222222222222e-05,
|
| 539 |
+
"loss": 0.3357,
|
| 540 |
"step": 840
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 3.79,
|
| 544 |
+
"learning_rate": 3.4474206349206354e-05,
|
| 545 |
+
"loss": 0.3738,
|
| 546 |
"step": 850
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 3.83,
|
| 550 |
+
"learning_rate": 3.422619047619048e-05,
|
| 551 |
+
"loss": 0.3769,
|
| 552 |
"step": 860
|
| 553 |
},
|
| 554 |
{
|
| 555 |
"epoch": 3.88,
|
| 556 |
+
"learning_rate": 3.397817460317461e-05,
|
| 557 |
+
"loss": 0.3667,
|
| 558 |
"step": 870
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 3.92,
|
| 562 |
+
"learning_rate": 3.3730158730158734e-05,
|
| 563 |
+
"loss": 0.3459,
|
| 564 |
"step": 880
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 3.96,
|
| 568 |
+
"learning_rate": 3.348214285714286e-05,
|
| 569 |
+
"loss": 0.3344,
|
| 570 |
"step": 890
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 4.0,
|
| 574 |
+
"eval_accuracy": 0.6989412092504876,
|
| 575 |
+
"eval_loss": 0.939809262752533,
|
| 576 |
+
"eval_runtime": 60.4275,
|
| 577 |
+
"eval_samples_per_second": 118.787,
|
| 578 |
+
"eval_steps_per_second": 3.723,
|
| 579 |
"step": 898
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 4.01,
|
| 583 |
+
"learning_rate": 3.3234126984126983e-05,
|
| 584 |
+
"loss": 0.3389,
|
| 585 |
"step": 900
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 4.05,
|
| 589 |
+
"learning_rate": 3.2986111111111115e-05,
|
| 590 |
+
"loss": 0.274,
|
| 591 |
"step": 910
|
| 592 |
},
|
| 593 |
{
|
| 594 |
"epoch": 4.1,
|
| 595 |
+
"learning_rate": 3.273809523809524e-05,
|
| 596 |
+
"loss": 0.2425,
|
| 597 |
"step": 920
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 4.14,
|
| 601 |
+
"learning_rate": 3.249007936507937e-05,
|
| 602 |
+
"loss": 0.2447,
|
| 603 |
"step": 930
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 4.19,
|
| 607 |
+
"learning_rate": 3.2242063492063495e-05,
|
| 608 |
+
"loss": 0.2604,
|
| 609 |
"step": 940
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 4.23,
|
| 613 |
+
"learning_rate": 3.199404761904762e-05,
|
| 614 |
+
"loss": 0.26,
|
| 615 |
"step": 950
|
| 616 |
},
|
| 617 |
{
|
| 618 |
"epoch": 4.28,
|
| 619 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 620 |
+
"loss": 0.2547,
|
| 621 |
"step": 960
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 4.32,
|
| 625 |
+
"learning_rate": 3.1498015873015876e-05,
|
| 626 |
+
"loss": 0.2642,
|
| 627 |
"step": 970
|
| 628 |
},
|
| 629 |
{
|
| 630 |
"epoch": 4.37,
|
| 631 |
+
"learning_rate": 3.125e-05,
|
| 632 |
+
"loss": 0.2564,
|
| 633 |
"step": 980
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 4.41,
|
| 637 |
+
"learning_rate": 3.100198412698413e-05,
|
| 638 |
+
"loss": 0.2667,
|
| 639 |
"step": 990
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 4.45,
|
| 643 |
+
"learning_rate": 3.075396825396826e-05,
|
| 644 |
+
"loss": 0.2687,
|
| 645 |
"step": 1000
|
| 646 |
},
|
| 647 |
{
|
| 648 |
"epoch": 4.5,
|
| 649 |
+
"learning_rate": 3.0505952380952385e-05,
|
| 650 |
+
"loss": 0.2744,
|
| 651 |
"step": 1010
|
| 652 |
},
|
| 653 |
{
|
| 654 |
"epoch": 4.54,
|
| 655 |
+
"learning_rate": 3.0257936507936506e-05,
|
| 656 |
+
"loss": 0.2594,
|
| 657 |
"step": 1020
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 4.59,
|
| 661 |
+
"learning_rate": 3.0009920634920634e-05,
|
| 662 |
+
"loss": 0.3404,
|
| 663 |
"step": 1030
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 4.63,
|
| 667 |
+
"learning_rate": 2.9761904761904762e-05,
|
| 668 |
+
"loss": 0.3398,
|
| 669 |
"step": 1040
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 4.68,
|
| 673 |
+
"learning_rate": 2.951388888888889e-05,
|
| 674 |
+
"loss": 0.2875,
|
| 675 |
"step": 1050
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 4.72,
|
| 679 |
+
"learning_rate": 2.9265873015873018e-05,
|
| 680 |
+
"loss": 0.3213,
|
| 681 |
"step": 1060
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 4.77,
|
| 685 |
+
"learning_rate": 2.9017857142857146e-05,
|
| 686 |
+
"loss": 0.3015,
|
| 687 |
"step": 1070
|
| 688 |
},
|
| 689 |
{
|
| 690 |
"epoch": 4.81,
|
| 691 |
+
"learning_rate": 2.876984126984127e-05,
|
| 692 |
+
"loss": 0.2809,
|
| 693 |
"step": 1080
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 4.86,
|
| 697 |
+
"learning_rate": 2.8521825396825395e-05,
|
| 698 |
+
"loss": 0.3107,
|
| 699 |
"step": 1090
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 4.9,
|
| 703 |
+
"learning_rate": 2.8273809523809523e-05,
|
| 704 |
+
"loss": 0.3181,
|
| 705 |
"step": 1100
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 4.94,
|
| 709 |
+
"learning_rate": 2.802579365079365e-05,
|
| 710 |
+
"loss": 0.2677,
|
| 711 |
"step": 1110
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 4.99,
|
| 715 |
+
"learning_rate": 2.777777777777778e-05,
|
| 716 |
+
"loss": 0.3055,
|
| 717 |
"step": 1120
|
| 718 |
},
|
| 719 |
{
|
| 720 |
+
"epoch": 5.0,
|
| 721 |
+
"eval_accuracy": 0.7057676232933965,
|
| 722 |
+
"eval_loss": 0.9803113341331482,
|
| 723 |
+
"eval_runtime": 60.4436,
|
| 724 |
+
"eval_samples_per_second": 118.755,
|
| 725 |
+
"eval_steps_per_second": 3.722,
|
| 726 |
+
"step": 1122
|
| 727 |
},
|
| 728 |
{
|
| 729 |
+
"epoch": 5.03,
|
| 730 |
+
"learning_rate": 2.7529761904761907e-05,
|
| 731 |
+
"loss": 0.2474,
|
| 732 |
+
"step": 1130
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 5.08,
|
| 736 |
+
"learning_rate": 2.7281746031746032e-05,
|
| 737 |
+
"loss": 0.2207,
|
| 738 |
+
"step": 1140
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 5.12,
|
| 742 |
+
"learning_rate": 2.703373015873016e-05,
|
| 743 |
+
"loss": 0.2116,
|
| 744 |
+
"step": 1150
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 5.17,
|
| 748 |
+
"learning_rate": 2.6785714285714288e-05,
|
| 749 |
+
"loss": 0.1894,
|
| 750 |
+
"step": 1160
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 5.21,
|
| 754 |
+
"learning_rate": 2.6537698412698416e-05,
|
| 755 |
+
"loss": 0.2606,
|
| 756 |
+
"step": 1170
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 5.26,
|
| 760 |
+
"learning_rate": 2.628968253968254e-05,
|
| 761 |
+
"loss": 0.1872,
|
| 762 |
+
"step": 1180
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 5.3,
|
| 766 |
+
"learning_rate": 2.604166666666667e-05,
|
| 767 |
+
"loss": 0.215,
|
| 768 |
+
"step": 1190
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 5.35,
|
| 772 |
+
"learning_rate": 2.5793650793650796e-05,
|
| 773 |
+
"loss": 0.1991,
|
| 774 |
+
"step": 1200
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 5.39,
|
| 778 |
+
"learning_rate": 2.554563492063492e-05,
|
| 779 |
+
"loss": 0.2105,
|
| 780 |
+
"step": 1210
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 5.43,
|
| 784 |
+
"learning_rate": 2.529761904761905e-05,
|
| 785 |
+
"loss": 0.2384,
|
| 786 |
+
"step": 1220
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 5.48,
|
| 790 |
+
"learning_rate": 2.5049603174603177e-05,
|
| 791 |
+
"loss": 0.222,
|
| 792 |
+
"step": 1230
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 5.52,
|
| 796 |
+
"learning_rate": 2.4801587301587305e-05,
|
| 797 |
+
"loss": 0.2175,
|
| 798 |
+
"step": 1240
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 5.57,
|
| 802 |
+
"learning_rate": 2.455357142857143e-05,
|
| 803 |
+
"loss": 0.2362,
|
| 804 |
+
"step": 1250
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 5.61,
|
| 808 |
+
"learning_rate": 2.4305555555555558e-05,
|
| 809 |
+
"loss": 0.2768,
|
| 810 |
+
"step": 1260
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 5.66,
|
| 814 |
+
"learning_rate": 2.4057539682539686e-05,
|
| 815 |
+
"loss": 0.2614,
|
| 816 |
+
"step": 1270
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 5.7,
|
| 820 |
+
"learning_rate": 2.380952380952381e-05,
|
| 821 |
+
"loss": 0.2127,
|
| 822 |
+
"step": 1280
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 5.75,
|
| 826 |
+
"learning_rate": 2.3561507936507938e-05,
|
| 827 |
+
"loss": 0.2232,
|
| 828 |
+
"step": 1290
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 5.79,
|
| 832 |
+
"learning_rate": 2.3313492063492066e-05,
|
| 833 |
+
"loss": 0.2089,
|
| 834 |
+
"step": 1300
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 5.84,
|
| 838 |
+
"learning_rate": 2.3065476190476194e-05,
|
| 839 |
+
"loss": 0.2321,
|
| 840 |
+
"step": 1310
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 5.88,
|
| 844 |
+
"learning_rate": 2.281746031746032e-05,
|
| 845 |
+
"loss": 0.242,
|
| 846 |
+
"step": 1320
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 5.92,
|
| 850 |
+
"learning_rate": 2.2569444444444447e-05,
|
| 851 |
+
"loss": 0.2388,
|
| 852 |
+
"step": 1330
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 5.97,
|
| 856 |
+
"learning_rate": 2.2321428571428575e-05,
|
| 857 |
+
"loss": 0.2214,
|
| 858 |
+
"step": 1340
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 6.0,
|
| 862 |
+
"eval_accuracy": 0.6953190303705767,
|
| 863 |
+
"eval_loss": 1.0336716175079346,
|
| 864 |
+
"eval_runtime": 60.3923,
|
| 865 |
+
"eval_samples_per_second": 118.856,
|
| 866 |
+
"eval_steps_per_second": 3.726,
|
| 867 |
+
"step": 1347
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 6.01,
|
| 871 |
+
"learning_rate": 2.20734126984127e-05,
|
| 872 |
+
"loss": 0.2192,
|
| 873 |
+
"step": 1350
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 6.06,
|
| 877 |
+
"learning_rate": 2.1825396825396827e-05,
|
| 878 |
+
"loss": 0.1988,
|
| 879 |
+
"step": 1360
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 6.1,
|
| 883 |
+
"learning_rate": 2.1577380952380955e-05,
|
| 884 |
+
"loss": 0.1624,
|
| 885 |
+
"step": 1370
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 6.15,
|
| 889 |
+
"learning_rate": 2.132936507936508e-05,
|
| 890 |
+
"loss": 0.1565,
|
| 891 |
+
"step": 1380
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 6.19,
|
| 895 |
+
"learning_rate": 2.1081349206349208e-05,
|
| 896 |
+
"loss": 0.1683,
|
| 897 |
+
"step": 1390
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 6.24,
|
| 901 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 902 |
+
"loss": 0.181,
|
| 903 |
+
"step": 1400
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 6.28,
|
| 907 |
+
"learning_rate": 2.058531746031746e-05,
|
| 908 |
+
"loss": 0.2063,
|
| 909 |
+
"step": 1410
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 6.33,
|
| 913 |
+
"learning_rate": 2.033730158730159e-05,
|
| 914 |
+
"loss": 0.1847,
|
| 915 |
+
"step": 1420
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 6.37,
|
| 919 |
+
"learning_rate": 2.0089285714285717e-05,
|
| 920 |
+
"loss": 0.1802,
|
| 921 |
+
"step": 1430
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 6.41,
|
| 925 |
+
"learning_rate": 1.984126984126984e-05,
|
| 926 |
+
"loss": 0.1626,
|
| 927 |
+
"step": 1440
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 6.46,
|
| 931 |
+
"learning_rate": 1.959325396825397e-05,
|
| 932 |
+
"loss": 0.1446,
|
| 933 |
+
"step": 1450
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 6.5,
|
| 937 |
+
"learning_rate": 1.9345238095238097e-05,
|
| 938 |
+
"loss": 0.1693,
|
| 939 |
+
"step": 1460
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 6.55,
|
| 943 |
+
"learning_rate": 1.9097222222222222e-05,
|
| 944 |
+
"loss": 0.1996,
|
| 945 |
+
"step": 1470
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 6.59,
|
| 949 |
+
"learning_rate": 1.884920634920635e-05,
|
| 950 |
+
"loss": 0.1653,
|
| 951 |
+
"step": 1480
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 6.64,
|
| 955 |
+
"learning_rate": 1.8601190476190478e-05,
|
| 956 |
+
"loss": 0.1622,
|
| 957 |
+
"step": 1490
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 6.68,
|
| 961 |
+
"learning_rate": 1.8353174603174602e-05,
|
| 962 |
+
"loss": 0.1654,
|
| 963 |
+
"step": 1500
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 6.73,
|
| 967 |
+
"learning_rate": 1.810515873015873e-05,
|
| 968 |
+
"loss": 0.1843,
|
| 969 |
+
"step": 1510
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 6.77,
|
| 973 |
+
"learning_rate": 1.785714285714286e-05,
|
| 974 |
+
"loss": 0.1779,
|
| 975 |
+
"step": 1520
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 6.82,
|
| 979 |
+
"learning_rate": 1.7609126984126986e-05,
|
| 980 |
+
"loss": 0.1801,
|
| 981 |
+
"step": 1530
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 6.86,
|
| 985 |
+
"learning_rate": 1.736111111111111e-05,
|
| 986 |
+
"loss": 0.1829,
|
| 987 |
+
"step": 1540
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 6.9,
|
| 991 |
+
"learning_rate": 1.711309523809524e-05,
|
| 992 |
+
"loss": 0.1789,
|
| 993 |
+
"step": 1550
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 6.95,
|
| 997 |
+
"learning_rate": 1.6865079365079367e-05,
|
| 998 |
+
"loss": 0.1894,
|
| 999 |
+
"step": 1560
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 6.99,
|
| 1003 |
+
"learning_rate": 1.6617063492063492e-05,
|
| 1004 |
+
"loss": 0.1575,
|
| 1005 |
+
"step": 1570
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 7.0,
|
| 1009 |
+
"eval_accuracy": 0.6976873780997492,
|
| 1010 |
+
"eval_loss": 1.0642220973968506,
|
| 1011 |
+
"eval_runtime": 59.9724,
|
| 1012 |
+
"eval_samples_per_second": 119.688,
|
| 1013 |
+
"eval_steps_per_second": 3.752,
|
| 1014 |
+
"step": 1571
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 7.04,
|
| 1018 |
+
"learning_rate": 1.636904761904762e-05,
|
| 1019 |
+
"loss": 0.1495,
|
| 1020 |
+
"step": 1580
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 7.08,
|
| 1024 |
+
"learning_rate": 1.6121031746031748e-05,
|
| 1025 |
+
"loss": 0.1392,
|
| 1026 |
+
"step": 1590
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 7.13,
|
| 1030 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 1031 |
+
"loss": 0.1095,
|
| 1032 |
+
"step": 1600
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 7.17,
|
| 1036 |
+
"learning_rate": 1.5625e-05,
|
| 1037 |
+
"loss": 0.1353,
|
| 1038 |
+
"step": 1610
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 7.22,
|
| 1042 |
+
"learning_rate": 1.537698412698413e-05,
|
| 1043 |
+
"loss": 0.1332,
|
| 1044 |
+
"step": 1620
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 7.26,
|
| 1048 |
+
"learning_rate": 1.5128968253968253e-05,
|
| 1049 |
+
"loss": 0.1227,
|
| 1050 |
+
"step": 1630
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 7.31,
|
| 1054 |
+
"learning_rate": 1.4880952380952381e-05,
|
| 1055 |
+
"loss": 0.1459,
|
| 1056 |
+
"step": 1640
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 7.35,
|
| 1060 |
+
"learning_rate": 1.4632936507936509e-05,
|
| 1061 |
+
"loss": 0.1221,
|
| 1062 |
+
"step": 1650
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 7.39,
|
| 1066 |
+
"learning_rate": 1.4384920634920635e-05,
|
| 1067 |
+
"loss": 0.1167,
|
| 1068 |
+
"step": 1660
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 7.44,
|
| 1072 |
+
"learning_rate": 1.4136904761904762e-05,
|
| 1073 |
+
"loss": 0.1597,
|
| 1074 |
+
"step": 1670
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 7.48,
|
| 1078 |
+
"learning_rate": 1.388888888888889e-05,
|
| 1079 |
+
"loss": 0.1395,
|
| 1080 |
+
"step": 1680
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 7.53,
|
| 1084 |
+
"learning_rate": 1.3640873015873016e-05,
|
| 1085 |
+
"loss": 0.1197,
|
| 1086 |
+
"step": 1690
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 7.57,
|
| 1090 |
+
"learning_rate": 1.3392857142857144e-05,
|
| 1091 |
+
"loss": 0.1435,
|
| 1092 |
+
"step": 1700
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 7.62,
|
| 1096 |
+
"learning_rate": 1.314484126984127e-05,
|
| 1097 |
+
"loss": 0.1084,
|
| 1098 |
+
"step": 1710
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 7.66,
|
| 1102 |
+
"learning_rate": 1.2896825396825398e-05,
|
| 1103 |
+
"loss": 0.1581,
|
| 1104 |
+
"step": 1720
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 7.71,
|
| 1108 |
+
"learning_rate": 1.2648809523809524e-05,
|
| 1109 |
+
"loss": 0.1389,
|
| 1110 |
+
"step": 1730
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 7.75,
|
| 1114 |
+
"learning_rate": 1.2400793650793652e-05,
|
| 1115 |
+
"loss": 0.1297,
|
| 1116 |
+
"step": 1740
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 7.8,
|
| 1120 |
+
"learning_rate": 1.2152777777777779e-05,
|
| 1121 |
+
"loss": 0.1641,
|
| 1122 |
+
"step": 1750
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 7.84,
|
| 1126 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 1127 |
+
"loss": 0.127,
|
| 1128 |
+
"step": 1760
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 7.88,
|
| 1132 |
+
"learning_rate": 1.1656746031746033e-05,
|
| 1133 |
+
"loss": 0.1252,
|
| 1134 |
+
"step": 1770
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 7.93,
|
| 1138 |
+
"learning_rate": 1.140873015873016e-05,
|
| 1139 |
+
"loss": 0.1661,
|
| 1140 |
+
"step": 1780
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 7.97,
|
| 1144 |
+
"learning_rate": 1.1160714285714287e-05,
|
| 1145 |
+
"loss": 0.1169,
|
| 1146 |
+
"step": 1790
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 8.0,
|
| 1150 |
+
"eval_accuracy": 0.7029813318473113,
|
| 1151 |
+
"eval_loss": 1.0828742980957031,
|
| 1152 |
+
"eval_runtime": 59.6867,
|
| 1153 |
+
"eval_samples_per_second": 120.261,
|
| 1154 |
+
"eval_steps_per_second": 3.77,
|
| 1155 |
+
"step": 1796
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 8.02,
|
| 1159 |
+
"learning_rate": 1.0912698412698414e-05,
|
| 1160 |
+
"loss": 0.1218,
|
| 1161 |
+
"step": 1800
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 8.06,
|
| 1165 |
+
"learning_rate": 1.066468253968254e-05,
|
| 1166 |
+
"loss": 0.1021,
|
| 1167 |
+
"step": 1810
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 8.11,
|
| 1171 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 1172 |
+
"loss": 0.1241,
|
| 1173 |
+
"step": 1820
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 8.15,
|
| 1177 |
+
"learning_rate": 1.0168650793650794e-05,
|
| 1178 |
+
"loss": 0.1004,
|
| 1179 |
+
"step": 1830
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 8.2,
|
| 1183 |
+
"learning_rate": 9.92063492063492e-06,
|
| 1184 |
+
"loss": 0.1279,
|
| 1185 |
+
"step": 1840
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 8.24,
|
| 1189 |
+
"learning_rate": 9.672619047619049e-06,
|
| 1190 |
+
"loss": 0.1132,
|
| 1191 |
+
"step": 1850
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 8.29,
|
| 1195 |
+
"learning_rate": 9.424603174603175e-06,
|
| 1196 |
+
"loss": 0.1137,
|
| 1197 |
+
"step": 1860
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 8.33,
|
| 1201 |
+
"learning_rate": 9.176587301587301e-06,
|
| 1202 |
+
"loss": 0.0887,
|
| 1203 |
+
"step": 1870
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 8.37,
|
| 1207 |
+
"learning_rate": 8.92857142857143e-06,
|
| 1208 |
+
"loss": 0.1115,
|
| 1209 |
+
"step": 1880
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 8.42,
|
| 1213 |
+
"learning_rate": 8.680555555555556e-06,
|
| 1214 |
+
"loss": 0.0955,
|
| 1215 |
+
"step": 1890
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 8.46,
|
| 1219 |
+
"learning_rate": 8.432539682539684e-06,
|
| 1220 |
+
"loss": 0.1268,
|
| 1221 |
+
"step": 1900
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 8.51,
|
| 1225 |
+
"learning_rate": 8.18452380952381e-06,
|
| 1226 |
+
"loss": 0.1007,
|
| 1227 |
+
"step": 1910
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 8.55,
|
| 1231 |
+
"learning_rate": 7.936507936507936e-06,
|
| 1232 |
+
"loss": 0.1025,
|
| 1233 |
+
"step": 1920
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 8.6,
|
| 1237 |
+
"learning_rate": 7.688492063492064e-06,
|
| 1238 |
+
"loss": 0.1082,
|
| 1239 |
+
"step": 1930
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 8.64,
|
| 1243 |
+
"learning_rate": 7.4404761904761905e-06,
|
| 1244 |
+
"loss": 0.0988,
|
| 1245 |
+
"step": 1940
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 8.69,
|
| 1249 |
+
"learning_rate": 7.192460317460318e-06,
|
| 1250 |
+
"loss": 0.1012,
|
| 1251 |
+
"step": 1950
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 8.73,
|
| 1255 |
+
"learning_rate": 6.944444444444445e-06,
|
| 1256 |
+
"loss": 0.1512,
|
| 1257 |
+
"step": 1960
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 8.78,
|
| 1261 |
+
"learning_rate": 6.696428571428572e-06,
|
| 1262 |
+
"loss": 0.0984,
|
| 1263 |
+
"step": 1970
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 8.82,
|
| 1267 |
+
"learning_rate": 6.448412698412699e-06,
|
| 1268 |
+
"loss": 0.1254,
|
| 1269 |
+
"step": 1980
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 8.86,
|
| 1273 |
+
"learning_rate": 6.200396825396826e-06,
|
| 1274 |
+
"loss": 0.0763,
|
| 1275 |
+
"step": 1990
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 8.91,
|
| 1279 |
+
"learning_rate": 5.9523809523809525e-06,
|
| 1280 |
+
"loss": 0.0746,
|
| 1281 |
+
"step": 2000
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 8.95,
|
| 1285 |
+
"learning_rate": 5.70436507936508e-06,
|
| 1286 |
+
"loss": 0.0933,
|
| 1287 |
+
"step": 2010
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 9.0,
|
| 1291 |
+
"learning_rate": 5.456349206349207e-06,
|
| 1292 |
+
"loss": 0.0917,
|
| 1293 |
+
"step": 2020
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 9.0,
|
| 1297 |
+
"eval_accuracy": 0.7047924212872666,
|
| 1298 |
+
"eval_loss": 1.1121457815170288,
|
| 1299 |
+
"eval_runtime": 59.8317,
|
| 1300 |
+
"eval_samples_per_second": 119.97,
|
| 1301 |
+
"eval_steps_per_second": 3.761,
|
| 1302 |
+
"step": 2020
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 9.04,
|
| 1306 |
+
"learning_rate": 5.208333333333334e-06,
|
| 1307 |
+
"loss": 0.078,
|
| 1308 |
+
"step": 2030
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 9.09,
|
| 1312 |
+
"learning_rate": 4.96031746031746e-06,
|
| 1313 |
+
"loss": 0.0989,
|
| 1314 |
+
"step": 2040
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 9.13,
|
| 1318 |
+
"learning_rate": 4.7123015873015875e-06,
|
| 1319 |
+
"loss": 0.0725,
|
| 1320 |
+
"step": 2050
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 9.18,
|
| 1324 |
+
"learning_rate": 4.464285714285715e-06,
|
| 1325 |
+
"loss": 0.0887,
|
| 1326 |
+
"step": 2060
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 9.22,
|
| 1330 |
+
"learning_rate": 4.216269841269842e-06,
|
| 1331 |
+
"loss": 0.0723,
|
| 1332 |
+
"step": 2070
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 9.27,
|
| 1336 |
+
"learning_rate": 3.968253968253968e-06,
|
| 1337 |
+
"loss": 0.0656,
|
| 1338 |
+
"step": 2080
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 9.31,
|
| 1342 |
+
"learning_rate": 3.7202380952380952e-06,
|
| 1343 |
+
"loss": 0.078,
|
| 1344 |
+
"step": 2090
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 9.35,
|
| 1348 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 1349 |
+
"loss": 0.1117,
|
| 1350 |
+
"step": 2100
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 9.4,
|
| 1354 |
+
"learning_rate": 3.2242063492063495e-06,
|
| 1355 |
+
"loss": 0.0865,
|
| 1356 |
+
"step": 2110
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 9.44,
|
| 1360 |
+
"learning_rate": 2.9761904761904763e-06,
|
| 1361 |
+
"loss": 0.1098,
|
| 1362 |
+
"step": 2120
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 9.49,
|
| 1366 |
+
"learning_rate": 2.7281746031746034e-06,
|
| 1367 |
+
"loss": 0.0955,
|
| 1368 |
+
"step": 2130
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 9.53,
|
| 1372 |
+
"learning_rate": 2.48015873015873e-06,
|
| 1373 |
+
"loss": 0.0943,
|
| 1374 |
+
"step": 2140
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 9.58,
|
| 1378 |
+
"learning_rate": 2.2321428571428573e-06,
|
| 1379 |
+
"loss": 0.0811,
|
| 1380 |
+
"step": 2150
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 9.62,
|
| 1384 |
+
"learning_rate": 1.984126984126984e-06,
|
| 1385 |
+
"loss": 0.0648,
|
| 1386 |
+
"step": 2160
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 9.67,
|
| 1390 |
+
"learning_rate": 1.7361111111111112e-06,
|
| 1391 |
+
"loss": 0.1009,
|
| 1392 |
+
"step": 2170
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 9.71,
|
| 1396 |
+
"learning_rate": 1.4880952380952381e-06,
|
| 1397 |
+
"loss": 0.082,
|
| 1398 |
+
"step": 2180
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 9.76,
|
| 1402 |
+
"learning_rate": 1.240079365079365e-06,
|
| 1403 |
+
"loss": 0.0714,
|
| 1404 |
+
"step": 2190
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 9.8,
|
| 1408 |
+
"learning_rate": 9.92063492063492e-07,
|
| 1409 |
+
"loss": 0.0613,
|
| 1410 |
+
"step": 2200
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 9.84,
|
| 1414 |
+
"learning_rate": 7.440476190476191e-07,
|
| 1415 |
+
"loss": 0.106,
|
| 1416 |
+
"step": 2210
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 9.89,
|
| 1420 |
+
"learning_rate": 4.96031746031746e-07,
|
| 1421 |
+
"loss": 0.0915,
|
| 1422 |
+
"step": 2220
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 9.93,
|
| 1426 |
+
"learning_rate": 2.48015873015873e-07,
|
| 1427 |
+
"loss": 0.0756,
|
| 1428 |
+
"step": 2230
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 9.98,
|
| 1432 |
+
"learning_rate": 0.0,
|
| 1433 |
+
"loss": 0.0785,
|
| 1434 |
+
"step": 2240
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 9.98,
|
| 1438 |
+
"eval_accuracy": 0.7052103650041794,
|
| 1439 |
+
"eval_loss": 1.1279975175857544,
|
| 1440 |
+
"eval_runtime": 66.0405,
|
| 1441 |
+
"eval_samples_per_second": 108.691,
|
| 1442 |
+
"eval_steps_per_second": 3.407,
|
| 1443 |
+
"step": 2240
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 9.98,
|
| 1447 |
+
"step": 2240,
|
| 1448 |
+
"total_flos": 2.2200667552042852e+19,
|
| 1449 |
+
"train_loss": 0.27600424638284105,
|
| 1450 |
+
"train_runtime": 6453.2956,
|
| 1451 |
+
"train_samples_per_second": 44.487,
|
| 1452 |
+
"train_steps_per_second": 0.347
|
| 1453 |
}
|
| 1454 |
],
|
| 1455 |
"logging_steps": 10,
|
| 1456 |
+
"max_steps": 2240,
|
| 1457 |
"num_input_tokens_seen": 0,
|
| 1458 |
+
"num_train_epochs": 10,
|
| 1459 |
"save_steps": 500,
|
| 1460 |
+
"total_flos": 2.2200667552042852e+19,
|
| 1461 |
"train_batch_size": 32,
|
| 1462 |
"trial_name": null,
|
| 1463 |
"trial_params": null
|