🍻 cheers
Browse files- README.md +3 -2
- all_results.json +10 -10
- eval_results.json +5 -5
- runs/Apr30_14-52-00_e47bc2c7ac19/events.out.tfevents.1714489523.e47bc2c7ac19.34.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +739 -455
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- imagefolder
|
|
@@ -14,7 +15,7 @@ model-index:
|
|
| 14 |
name: Image Classification
|
| 15 |
type: image-classification
|
| 16 |
dataset:
|
| 17 |
-
name:
|
| 18 |
type: imagefolder
|
| 19 |
config: default
|
| 20 |
split: train
|
|
@@ -30,7 +31,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# Action_all_10_class
|
| 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
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
- Loss: 0.4146
|
| 36 |
- Accuracy: 0.8785
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- imagefolder
|
|
|
|
| 15 |
name: Image Classification
|
| 16 |
type: image-classification
|
| 17 |
dataset:
|
| 18 |
+
name: Action_small_dataset
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: train
|
|
|
|
| 31 |
|
| 32 |
# Action_all_10_class
|
| 33 |
|
| 34 |
+
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 Action_small_dataset dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
- Loss: 0.4146
|
| 37 |
- Accuracy: 0.8785
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 11.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos": 1.
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.8785276073619632,
|
| 4 |
+
"eval_loss": 0.4145749807357788,
|
| 5 |
+
"eval_runtime": 11.5662,
|
| 6 |
+
"eval_samples_per_second": 70.464,
|
| 7 |
+
"eval_steps_per_second": 8.819,
|
| 8 |
+
"total_flos": 1.789030847196795e+18,
|
| 9 |
+
"train_loss": 0.6435174308433664,
|
| 10 |
+
"train_runtime": 776.1745,
|
| 11 |
+
"train_samples_per_second": 29.742,
|
| 12 |
+
"train_steps_per_second": 1.862
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 11.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.8785276073619632,
|
| 4 |
+
"eval_loss": 0.4145749807357788,
|
| 5 |
+
"eval_runtime": 11.5662,
|
| 6 |
+
"eval_samples_per_second": 70.464,
|
| 7 |
+
"eval_steps_per_second": 8.819
|
| 8 |
}
|
runs/Apr30_14-52-00_e47bc2c7ac19/events.out.tfevents.1714489523.e47bc2c7ac19.34.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b915e629320c5d37a2eb5c4eb05ad751b260745a4910295acff1688f81bc0ad7
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"total_flos": 1.
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.789030847196795e+18,
|
| 4 |
+
"train_loss": 0.6435174308433664,
|
| 5 |
+
"train_runtime": 776.1745,
|
| 6 |
+
"train_samples_per_second": 29.742,
|
| 7 |
+
"train_steps_per_second": 1.862
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,880 +1,1164 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "Action_all_10_class/checkpoint-
|
| 4 |
"epoch": 5.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"
|
| 14 |
-
"
|
|
|
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"
|
| 20 |
-
"
|
|
|
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"
|
| 26 |
-
"
|
|
|
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"
|
| 32 |
-
"
|
|
|
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"
|
| 38 |
-
"
|
|
|
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"
|
| 44 |
-
"
|
|
|
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"
|
| 50 |
-
"
|
|
|
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"
|
| 56 |
-
"
|
|
|
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
-
"
|
| 62 |
-
"
|
|
|
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch": 0.
|
| 67 |
-
"
|
| 68 |
-
"
|
|
|
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"eval_accuracy": 0.
|
| 74 |
-
"eval_loss":
|
| 75 |
-
"eval_runtime":
|
| 76 |
-
"eval_samples_per_second":
|
| 77 |
-
"eval_steps_per_second": 7.
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch": 0.
|
| 82 |
-
"
|
| 83 |
-
"
|
|
|
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch": 0.
|
| 88 |
-
"
|
| 89 |
-
"
|
|
|
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch": 0.
|
| 94 |
-
"
|
| 95 |
-
"
|
|
|
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch": 0.
|
| 100 |
-
"
|
| 101 |
-
"
|
|
|
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 0.
|
| 106 |
-
"
|
| 107 |
-
"
|
|
|
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch": 0.
|
| 112 |
-
"
|
| 113 |
-
"
|
|
|
|
| 114 |
"step": 160
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"
|
| 119 |
-
"
|
|
|
|
| 120 |
"step": 170
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch": 0.
|
| 124 |
-
"
|
| 125 |
-
"
|
|
|
|
| 126 |
"step": 180
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch": 0.
|
| 130 |
-
"
|
| 131 |
-
"
|
|
|
|
| 132 |
"step": 190
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch": 0.
|
| 136 |
-
"
|
| 137 |
-
"
|
|
|
|
| 138 |
"step": 200
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch": 0.
|
| 142 |
-
"eval_accuracy": 0.
|
| 143 |
-
"eval_loss": 0.
|
| 144 |
-
"eval_runtime":
|
| 145 |
-
"eval_samples_per_second":
|
| 146 |
-
"eval_steps_per_second":
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 0.
|
| 151 |
-
"
|
| 152 |
-
"
|
|
|
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 0.
|
| 157 |
-
"
|
| 158 |
-
"
|
|
|
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch": 0.
|
| 163 |
-
"
|
| 164 |
-
"
|
|
|
|
| 165 |
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch": 0.
|
| 169 |
-
"
|
| 170 |
-
"
|
|
|
|
| 171 |
"step": 240
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch":
|
| 175 |
-
"
|
| 176 |
-
"
|
|
|
|
| 177 |
"step": 250
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"
|
| 182 |
-
"
|
|
|
|
| 183 |
"step": 260
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"
|
| 188 |
-
"
|
|
|
|
| 189 |
"step": 270
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"
|
| 194 |
-
"
|
|
|
|
| 195 |
"step": 280
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 1.
|
| 199 |
-
"
|
| 200 |
-
"
|
|
|
|
| 201 |
"step": 290
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch": 1.
|
| 205 |
-
"
|
| 206 |
-
"
|
|
|
|
| 207 |
"step": 300
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 1.
|
| 211 |
-
"eval_accuracy": 0.
|
| 212 |
-
"eval_loss": 0.
|
| 213 |
-
"eval_runtime":
|
| 214 |
-
"eval_samples_per_second":
|
| 215 |
-
"eval_steps_per_second":
|
| 216 |
"step": 300
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch": 1.
|
| 220 |
-
"
|
| 221 |
-
"
|
|
|
|
| 222 |
"step": 310
|
| 223 |
},
|
| 224 |
{
|
| 225 |
-
"epoch": 1.
|
| 226 |
-
"
|
| 227 |
-
"
|
|
|
|
| 228 |
"step": 320
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch": 1.
|
| 232 |
-
"
|
| 233 |
-
"
|
|
|
|
| 234 |
"step": 330
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 1.
|
| 238 |
-
"
|
| 239 |
-
"
|
|
|
|
| 240 |
"step": 340
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 1.
|
| 244 |
-
"
|
| 245 |
-
"
|
|
|
|
| 246 |
"step": 350
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch": 1.
|
| 250 |
-
"
|
| 251 |
-
"
|
|
|
|
| 252 |
"step": 360
|
| 253 |
},
|
| 254 |
{
|
| 255 |
-
"epoch": 1.
|
| 256 |
-
"
|
| 257 |
-
"
|
|
|
|
| 258 |
"step": 370
|
| 259 |
},
|
| 260 |
{
|
| 261 |
-
"epoch": 1.
|
| 262 |
-
"
|
| 263 |
-
"
|
|
|
|
| 264 |
"step": 380
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch": 1.
|
| 268 |
-
"
|
| 269 |
-
"
|
|
|
|
| 270 |
"step": 390
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch": 1.
|
| 274 |
-
"
|
| 275 |
-
"
|
|
|
|
| 276 |
"step": 400
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch": 1.
|
| 280 |
-
"eval_accuracy": 0.
|
| 281 |
-
"eval_loss": 0.
|
| 282 |
-
"eval_runtime":
|
| 283 |
-
"eval_samples_per_second":
|
| 284 |
-
"eval_steps_per_second":
|
| 285 |
"step": 400
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch": 1.
|
| 289 |
-
"
|
| 290 |
-
"
|
|
|
|
| 291 |
"step": 410
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch": 1.
|
| 295 |
-
"
|
| 296 |
-
"
|
|
|
|
| 297 |
"step": 420
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 1.
|
| 301 |
-
"
|
| 302 |
-
"
|
|
|
|
| 303 |
"step": 430
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 1.
|
| 307 |
-
"
|
| 308 |
-
"
|
|
|
|
| 309 |
"step": 440
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch": 1.
|
| 313 |
-
"
|
| 314 |
-
"
|
|
|
|
| 315 |
"step": 450
|
| 316 |
},
|
| 317 |
{
|
| 318 |
-
"epoch": 1.
|
| 319 |
-
"
|
| 320 |
-
"
|
|
|
|
| 321 |
"step": 460
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch": 1.
|
| 325 |
-
"
|
| 326 |
-
"
|
|
|
|
| 327 |
"step": 470
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch": 1.
|
| 331 |
-
"
|
| 332 |
-
"
|
|
|
|
| 333 |
"step": 480
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch": 1.
|
| 337 |
-
"
|
| 338 |
-
"
|
|
|
|
| 339 |
"step": 490
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
-
"
|
| 344 |
-
"
|
|
|
|
| 345 |
"step": 500
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
-
"eval_accuracy": 0.
|
| 350 |
-
"eval_loss": 0.
|
| 351 |
-
"eval_runtime":
|
| 352 |
-
"eval_samples_per_second":
|
| 353 |
-
"eval_steps_per_second":
|
| 354 |
"step": 500
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch":
|
| 358 |
-
"
|
| 359 |
-
"
|
|
|
|
| 360 |
"step": 510
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
-
"
|
| 365 |
-
"
|
|
|
|
| 366 |
"step": 520
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch":
|
| 370 |
-
"
|
| 371 |
-
"
|
|
|
|
| 372 |
"step": 530
|
| 373 |
},
|
| 374 |
{
|
| 375 |
-
"epoch":
|
| 376 |
-
"
|
| 377 |
-
"
|
|
|
|
| 378 |
"step": 540
|
| 379 |
},
|
| 380 |
{
|
| 381 |
-
"epoch":
|
| 382 |
-
"
|
| 383 |
-
"
|
|
|
|
| 384 |
"step": 550
|
| 385 |
},
|
| 386 |
{
|
| 387 |
-
"epoch":
|
| 388 |
-
"
|
| 389 |
-
"
|
|
|
|
| 390 |
"step": 560
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"epoch":
|
| 394 |
-
"
|
| 395 |
-
"
|
|
|
|
| 396 |
"step": 570
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch": 2.
|
| 400 |
-
"
|
| 401 |
-
"
|
|
|
|
| 402 |
"step": 580
|
| 403 |
},
|
| 404 |
{
|
| 405 |
-
"epoch": 2.
|
| 406 |
-
"
|
| 407 |
-
"
|
|
|
|
| 408 |
"step": 590
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 2.
|
| 412 |
-
"
|
| 413 |
-
"
|
|
|
|
| 414 |
"step": 600
|
| 415 |
},
|
| 416 |
{
|
| 417 |
-
"epoch": 2.
|
| 418 |
-
"eval_accuracy": 0.
|
| 419 |
-
"eval_loss": 0.
|
| 420 |
-
"eval_runtime":
|
| 421 |
-
"eval_samples_per_second":
|
| 422 |
-
"eval_steps_per_second":
|
| 423 |
"step": 600
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch": 2.
|
| 427 |
-
"
|
| 428 |
-
"
|
|
|
|
| 429 |
"step": 610
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 2.
|
| 433 |
-
"
|
| 434 |
-
"
|
|
|
|
| 435 |
"step": 620
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch": 2.
|
| 439 |
-
"
|
| 440 |
-
"
|
|
|
|
| 441 |
"step": 630
|
| 442 |
},
|
| 443 |
{
|
| 444 |
-
"epoch": 2.
|
| 445 |
-
"
|
| 446 |
-
"
|
|
|
|
| 447 |
"step": 640
|
| 448 |
},
|
| 449 |
{
|
| 450 |
-
"epoch": 2.
|
| 451 |
-
"
|
| 452 |
-
"
|
|
|
|
| 453 |
"step": 650
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch": 2.
|
| 457 |
-
"
|
| 458 |
-
"
|
|
|
|
| 459 |
"step": 660
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch": 2.
|
| 463 |
-
"
|
| 464 |
-
"
|
|
|
|
| 465 |
"step": 670
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch": 2.
|
| 469 |
-
"
|
| 470 |
-
"
|
|
|
|
| 471 |
"step": 680
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 2.
|
| 475 |
-
"
|
| 476 |
-
"
|
|
|
|
| 477 |
"step": 690
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"epoch": 2.
|
| 481 |
-
"
|
| 482 |
-
"
|
|
|
|
| 483 |
"step": 700
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch": 2.
|
| 487 |
-
"eval_accuracy": 0.
|
| 488 |
-
"eval_loss": 0.
|
| 489 |
-
"eval_runtime":
|
| 490 |
-
"eval_samples_per_second":
|
| 491 |
-
"eval_steps_per_second":
|
| 492 |
"step": 700
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 2.
|
| 496 |
-
"
|
| 497 |
-
"
|
|
|
|
| 498 |
"step": 710
|
| 499 |
},
|
| 500 |
{
|
| 501 |
-
"epoch": 2.
|
| 502 |
-
"
|
| 503 |
-
"
|
|
|
|
| 504 |
"step": 720
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch": 2.
|
| 508 |
-
"
|
| 509 |
-
"
|
|
|
|
| 510 |
"step": 730
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch": 2.
|
| 514 |
-
"
|
| 515 |
-
"
|
|
|
|
| 516 |
"step": 740
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch":
|
| 520 |
-
"
|
| 521 |
-
"
|
|
|
|
| 522 |
"step": 750
|
| 523 |
},
|
| 524 |
{
|
| 525 |
-
"epoch":
|
| 526 |
-
"
|
| 527 |
-
"
|
|
|
|
| 528 |
"step": 760
|
| 529 |
},
|
| 530 |
{
|
| 531 |
-
"epoch":
|
| 532 |
-
"
|
| 533 |
-
"
|
|
|
|
| 534 |
"step": 770
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch":
|
| 538 |
-
"
|
| 539 |
-
"
|
|
|
|
| 540 |
"step": 780
|
| 541 |
},
|
| 542 |
{
|
| 543 |
-
"epoch":
|
| 544 |
-
"
|
| 545 |
-
"
|
|
|
|
| 546 |
"step": 790
|
| 547 |
},
|
| 548 |
{
|
| 549 |
-
"epoch":
|
| 550 |
-
"
|
| 551 |
-
"
|
|
|
|
| 552 |
"step": 800
|
| 553 |
},
|
| 554 |
{
|
| 555 |
-
"epoch":
|
| 556 |
-
"eval_accuracy": 0.
|
| 557 |
-
"eval_loss": 0.
|
| 558 |
-
"eval_runtime":
|
| 559 |
-
"eval_samples_per_second":
|
| 560 |
-
"eval_steps_per_second": 8.
|
| 561 |
"step": 800
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"epoch":
|
| 565 |
-
"
|
| 566 |
-
"
|
|
|
|
| 567 |
"step": 810
|
| 568 |
},
|
| 569 |
{
|
| 570 |
-
"epoch":
|
| 571 |
-
"
|
| 572 |
-
"
|
|
|
|
| 573 |
"step": 820
|
| 574 |
},
|
| 575 |
{
|
| 576 |
-
"epoch":
|
| 577 |
-
"
|
| 578 |
-
"
|
|
|
|
| 579 |
"step": 830
|
| 580 |
},
|
| 581 |
{
|
| 582 |
-
"epoch":
|
| 583 |
-
"
|
| 584 |
-
"
|
|
|
|
| 585 |
"step": 840
|
| 586 |
},
|
| 587 |
{
|
| 588 |
-
"epoch":
|
| 589 |
-
"
|
| 590 |
-
"
|
|
|
|
| 591 |
"step": 850
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch":
|
| 595 |
-
"
|
| 596 |
-
"
|
|
|
|
| 597 |
"step": 860
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 3.
|
| 601 |
-
"
|
| 602 |
-
"
|
|
|
|
| 603 |
"step": 870
|
| 604 |
},
|
| 605 |
{
|
| 606 |
-
"epoch": 3.
|
| 607 |
-
"
|
| 608 |
-
"
|
|
|
|
| 609 |
"step": 880
|
| 610 |
},
|
| 611 |
{
|
| 612 |
-
"epoch": 3.
|
| 613 |
-
"
|
| 614 |
-
"
|
|
|
|
| 615 |
"step": 890
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch": 3.
|
| 619 |
-
"
|
| 620 |
-
"
|
|
|
|
| 621 |
"step": 900
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"epoch": 3.
|
| 625 |
-
"eval_accuracy": 0.
|
| 626 |
-
"eval_loss": 0.
|
| 627 |
-
"eval_runtime": 11.
|
| 628 |
-
"eval_samples_per_second":
|
| 629 |
-
"eval_steps_per_second":
|
| 630 |
"step": 900
|
| 631 |
},
|
| 632 |
{
|
| 633 |
-
"epoch": 3.
|
| 634 |
-
"
|
| 635 |
-
"
|
|
|
|
| 636 |
"step": 910
|
| 637 |
},
|
| 638 |
{
|
| 639 |
-
"epoch": 3.
|
| 640 |
-
"
|
| 641 |
-
"
|
|
|
|
| 642 |
"step": 920
|
| 643 |
},
|
| 644 |
{
|
| 645 |
-
"epoch": 3.
|
| 646 |
-
"
|
| 647 |
-
"
|
|
|
|
| 648 |
"step": 930
|
| 649 |
},
|
| 650 |
{
|
| 651 |
-
"epoch": 3.
|
| 652 |
-
"
|
| 653 |
-
"
|
|
|
|
| 654 |
"step": 940
|
| 655 |
},
|
| 656 |
{
|
| 657 |
-
"epoch": 3.
|
| 658 |
-
"
|
| 659 |
-
"
|
|
|
|
| 660 |
"step": 950
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch": 3.
|
| 664 |
-
"
|
| 665 |
-
"
|
|
|
|
| 666 |
"step": 960
|
| 667 |
},
|
| 668 |
{
|
| 669 |
-
"epoch": 3.
|
| 670 |
-
"
|
| 671 |
-
"
|
|
|
|
| 672 |
"step": 970
|
| 673 |
},
|
| 674 |
{
|
| 675 |
-
"epoch": 3.
|
| 676 |
-
"
|
| 677 |
-
"
|
|
|
|
| 678 |
"step": 980
|
| 679 |
},
|
| 680 |
{
|
| 681 |
-
"epoch": 3.
|
| 682 |
-
"
|
| 683 |
-
"
|
|
|
|
| 684 |
"step": 990
|
| 685 |
},
|
| 686 |
{
|
| 687 |
-
"epoch":
|
| 688 |
-
"
|
| 689 |
-
"
|
|
|
|
| 690 |
"step": 1000
|
| 691 |
},
|
| 692 |
{
|
| 693 |
-
"epoch":
|
| 694 |
-
"eval_accuracy": 0.
|
| 695 |
-
"eval_loss": 0.
|
| 696 |
-
"eval_runtime": 11.
|
| 697 |
-
"eval_samples_per_second":
|
| 698 |
-
"eval_steps_per_second":
|
| 699 |
"step": 1000
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch":
|
| 703 |
-
"
|
| 704 |
-
"
|
|
|
|
| 705 |
"step": 1010
|
| 706 |
},
|
| 707 |
{
|
| 708 |
-
"epoch":
|
| 709 |
-
"
|
| 710 |
-
"
|
|
|
|
| 711 |
"step": 1020
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"epoch":
|
| 715 |
-
"
|
| 716 |
-
"
|
|
|
|
| 717 |
"step": 1030
|
| 718 |
},
|
| 719 |
{
|
| 720 |
-
"epoch":
|
| 721 |
-
"
|
| 722 |
-
"
|
|
|
|
| 723 |
"step": 1040
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"epoch":
|
| 727 |
-
"
|
| 728 |
-
"
|
|
|
|
| 729 |
"step": 1050
|
| 730 |
},
|
| 731 |
{
|
| 732 |
-
"epoch":
|
| 733 |
-
"
|
| 734 |
-
"
|
|
|
|
| 735 |
"step": 1060
|
| 736 |
},
|
| 737 |
{
|
| 738 |
-
"epoch":
|
| 739 |
-
"
|
| 740 |
-
"
|
|
|
|
| 741 |
"step": 1070
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch":
|
| 745 |
-
"
|
| 746 |
-
"
|
|
|
|
| 747 |
"step": 1080
|
| 748 |
},
|
| 749 |
{
|
| 750 |
-
"epoch":
|
| 751 |
-
"
|
| 752 |
-
"
|
|
|
|
| 753 |
"step": 1090
|
| 754 |
},
|
| 755 |
{
|
| 756 |
-
"epoch":
|
| 757 |
-
"
|
| 758 |
-
"
|
|
|
|
| 759 |
"step": 1100
|
| 760 |
},
|
| 761 |
{
|
| 762 |
-
"epoch":
|
| 763 |
-
"eval_accuracy": 0.
|
| 764 |
-
"eval_loss": 0.
|
| 765 |
-
"eval_runtime":
|
| 766 |
-
"eval_samples_per_second":
|
| 767 |
-
"eval_steps_per_second":
|
| 768 |
"step": 1100
|
| 769 |
},
|
| 770 |
{
|
| 771 |
-
"epoch":
|
| 772 |
-
"
|
| 773 |
-
"
|
|
|
|
| 774 |
"step": 1110
|
| 775 |
},
|
| 776 |
{
|
| 777 |
-
"epoch":
|
| 778 |
-
"
|
| 779 |
-
"
|
|
|
|
| 780 |
"step": 1120
|
| 781 |
},
|
| 782 |
{
|
| 783 |
-
"epoch":
|
| 784 |
-
"
|
| 785 |
-
"
|
|
|
|
| 786 |
"step": 1130
|
| 787 |
},
|
| 788 |
{
|
| 789 |
-
"epoch":
|
| 790 |
-
"
|
| 791 |
-
"
|
|
|
|
| 792 |
"step": 1140
|
| 793 |
},
|
| 794 |
{
|
| 795 |
-
"epoch":
|
| 796 |
-
"
|
| 797 |
-
"
|
|
|
|
| 798 |
"step": 1150
|
| 799 |
},
|
| 800 |
{
|
| 801 |
-
"epoch": 4.
|
| 802 |
-
"
|
| 803 |
-
"
|
|
|
|
| 804 |
"step": 1160
|
| 805 |
},
|
| 806 |
{
|
| 807 |
-
"epoch": 4.
|
| 808 |
-
"
|
| 809 |
-
"
|
|
|
|
| 810 |
"step": 1170
|
| 811 |
},
|
| 812 |
{
|
| 813 |
-
"epoch": 4.
|
| 814 |
-
"
|
| 815 |
-
"
|
|
|
|
| 816 |
"step": 1180
|
| 817 |
},
|
| 818 |
{
|
| 819 |
-
"epoch": 4.
|
| 820 |
-
"
|
| 821 |
-
"
|
|
|
|
| 822 |
"step": 1190
|
| 823 |
},
|
| 824 |
{
|
| 825 |
-
"epoch": 4.
|
| 826 |
-
"
|
| 827 |
-
"
|
|
|
|
| 828 |
"step": 1200
|
| 829 |
},
|
| 830 |
{
|
| 831 |
-
"epoch": 4.
|
| 832 |
-
"eval_accuracy": 0.
|
| 833 |
-
"eval_loss": 0.
|
| 834 |
-
"eval_runtime":
|
| 835 |
-
"eval_samples_per_second":
|
| 836 |
-
"eval_steps_per_second":
|
| 837 |
"step": 1200
|
| 838 |
},
|
| 839 |
{
|
| 840 |
-
"epoch": 4.
|
| 841 |
-
"
|
| 842 |
-
"
|
|
|
|
| 843 |
"step": 1210
|
| 844 |
},
|
| 845 |
{
|
| 846 |
-
"epoch": 4.
|
| 847 |
-
"
|
| 848 |
-
"
|
|
|
|
| 849 |
"step": 1220
|
| 850 |
},
|
| 851 |
{
|
| 852 |
-
"epoch": 4.
|
| 853 |
-
"
|
| 854 |
-
"
|
|
|
|
| 855 |
"step": 1230
|
| 856 |
},
|
| 857 |
{
|
| 858 |
-
"epoch": 4.
|
| 859 |
-
"
|
| 860 |
-
"
|
|
|
|
| 861 |
"step": 1240
|
| 862 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 863 |
{
|
| 864 |
"epoch": 5.0,
|
| 865 |
-
"step":
|
| 866 |
-
"total_flos": 1.
|
| 867 |
-
"train_loss": 0.
|
| 868 |
-
"train_runtime":
|
| 869 |
-
"train_samples_per_second":
|
| 870 |
-
"train_steps_per_second": 1.
|
| 871 |
}
|
| 872 |
],
|
| 873 |
"logging_steps": 10,
|
| 874 |
-
"max_steps":
|
|
|
|
| 875 |
"num_train_epochs": 5,
|
| 876 |
"save_steps": 100,
|
| 877 |
-
"total_flos": 1.
|
|
|
|
| 878 |
"trial_name": null,
|
| 879 |
"trial_params": null
|
| 880 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.4145749807357788,
|
| 3 |
+
"best_model_checkpoint": "Action_all_10_class/checkpoint-1400",
|
| 4 |
"epoch": 5.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 1445,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"grad_norm": 2.3712973594665527,
|
| 14 |
+
"learning_rate": 0.0001986159169550173,
|
| 15 |
+
"loss": 2.271,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.07,
|
| 20 |
+
"grad_norm": 1.8230082988739014,
|
| 21 |
+
"learning_rate": 0.0001972318339100346,
|
| 22 |
+
"loss": 2.0335,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.1,
|
| 27 |
+
"grad_norm": 2.621448040008545,
|
| 28 |
+
"learning_rate": 0.0001958477508650519,
|
| 29 |
+
"loss": 1.8885,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.14,
|
| 34 |
+
"grad_norm": 3.194303274154663,
|
| 35 |
+
"learning_rate": 0.0001944636678200692,
|
| 36 |
+
"loss": 1.6835,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.17,
|
| 41 |
+
"grad_norm": 2.9692325592041016,
|
| 42 |
+
"learning_rate": 0.0001930795847750865,
|
| 43 |
+
"loss": 1.5658,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.21,
|
| 48 |
+
"grad_norm": 3.6100592613220215,
|
| 49 |
+
"learning_rate": 0.00019169550173010383,
|
| 50 |
+
"loss": 1.4881,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.24,
|
| 55 |
+
"grad_norm": 2.926745653152466,
|
| 56 |
+
"learning_rate": 0.00019031141868512113,
|
| 57 |
+
"loss": 1.2581,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.28,
|
| 62 |
+
"grad_norm": 3.4733943939208984,
|
| 63 |
+
"learning_rate": 0.00018892733564013843,
|
| 64 |
+
"loss": 1.1786,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.31,
|
| 69 |
+
"grad_norm": 3.609900951385498,
|
| 70 |
+
"learning_rate": 0.00018754325259515573,
|
| 71 |
+
"loss": 1.1612,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.35,
|
| 76 |
+
"grad_norm": 3.1664867401123047,
|
| 77 |
+
"learning_rate": 0.00018615916955017303,
|
| 78 |
+
"loss": 1.1239,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.35,
|
| 83 |
+
"eval_accuracy": 0.7116564417177914,
|
| 84 |
+
"eval_loss": 0.9933902025222778,
|
| 85 |
+
"eval_runtime": 13.7579,
|
| 86 |
+
"eval_samples_per_second": 59.239,
|
| 87 |
+
"eval_steps_per_second": 7.414,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 0.38,
|
| 92 |
+
"grad_norm": 3.3514106273651123,
|
| 93 |
+
"learning_rate": 0.00018477508650519033,
|
| 94 |
+
"loss": 1.1763,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 0.42,
|
| 99 |
+
"grad_norm": 2.530064821243286,
|
| 100 |
+
"learning_rate": 0.0001833910034602076,
|
| 101 |
+
"loss": 1.1493,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 0.45,
|
| 106 |
+
"grad_norm": 4.615677833557129,
|
| 107 |
+
"learning_rate": 0.0001820069204152249,
|
| 108 |
+
"loss": 1.1661,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 0.48,
|
| 113 |
+
"grad_norm": 5.2710065841674805,
|
| 114 |
+
"learning_rate": 0.0001807612456747405,
|
| 115 |
+
"loss": 1.1174,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.52,
|
| 120 |
+
"grad_norm": 2.854151487350464,
|
| 121 |
+
"learning_rate": 0.0001793771626297578,
|
| 122 |
+
"loss": 1.0679,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 0.55,
|
| 127 |
+
"grad_norm": 4.68364143371582,
|
| 128 |
+
"learning_rate": 0.0001779930795847751,
|
| 129 |
+
"loss": 1.1566,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 0.59,
|
| 134 |
+
"grad_norm": 5.799492359161377,
|
| 135 |
+
"learning_rate": 0.0001766089965397924,
|
| 136 |
+
"loss": 0.9811,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 0.62,
|
| 141 |
+
"grad_norm": 7.8334760665893555,
|
| 142 |
+
"learning_rate": 0.0001752249134948097,
|
| 143 |
+
"loss": 1.0781,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 0.66,
|
| 148 |
+
"grad_norm": 4.817080020904541,
|
| 149 |
+
"learning_rate": 0.000173840830449827,
|
| 150 |
+
"loss": 1.0302,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
+
"epoch": 0.69,
|
| 155 |
+
"grad_norm": 4.084784507751465,
|
| 156 |
+
"learning_rate": 0.0001724567474048443,
|
| 157 |
+
"loss": 0.8878,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.69,
|
| 162 |
+
"eval_accuracy": 0.7705521472392638,
|
| 163 |
+
"eval_loss": 0.7667088508605957,
|
| 164 |
+
"eval_runtime": 10.9837,
|
| 165 |
+
"eval_samples_per_second": 74.201,
|
| 166 |
+
"eval_steps_per_second": 9.286,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 0.73,
|
| 171 |
+
"grad_norm": 3.8510003089904785,
|
| 172 |
+
"learning_rate": 0.0001710726643598616,
|
| 173 |
+
"loss": 1.0539,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 0.76,
|
| 178 |
+
"grad_norm": 4.4905619621276855,
|
| 179 |
+
"learning_rate": 0.00016968858131487892,
|
| 180 |
+
"loss": 0.9855,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 0.8,
|
| 185 |
+
"grad_norm": 3.903388738632202,
|
| 186 |
+
"learning_rate": 0.00016830449826989622,
|
| 187 |
+
"loss": 1.0442,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.83,
|
| 192 |
+
"grad_norm": 4.052041530609131,
|
| 193 |
+
"learning_rate": 0.00016692041522491352,
|
| 194 |
+
"loss": 1.004,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 0.87,
|
| 199 |
+
"grad_norm": 4.503437042236328,
|
| 200 |
+
"learning_rate": 0.00016553633217993081,
|
| 201 |
+
"loss": 0.8825,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 0.9,
|
| 206 |
+
"grad_norm": 4.2403459548950195,
|
| 207 |
+
"learning_rate": 0.00016415224913494811,
|
| 208 |
+
"loss": 0.9616,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 0.93,
|
| 213 |
+
"grad_norm": 6.690958499908447,
|
| 214 |
+
"learning_rate": 0.00016276816608996541,
|
| 215 |
+
"loss": 0.9127,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 0.97,
|
| 220 |
+
"grad_norm": 6.591899394989014,
|
| 221 |
+
"learning_rate": 0.0001613840830449827,
|
| 222 |
+
"loss": 1.0545,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 1.0,
|
| 227 |
+
"grad_norm": 2.814940929412842,
|
| 228 |
+
"learning_rate": 0.0001601384083044983,
|
| 229 |
+
"loss": 0.9937,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 1.04,
|
| 234 |
+
"grad_norm": 5.890912055969238,
|
| 235 |
+
"learning_rate": 0.00015875432525951557,
|
| 236 |
+
"loss": 0.9037,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 1.04,
|
| 241 |
+
"eval_accuracy": 0.8098159509202454,
|
| 242 |
+
"eval_loss": 0.6369422674179077,
|
| 243 |
+
"eval_runtime": 11.1962,
|
| 244 |
+
"eval_samples_per_second": 72.793,
|
| 245 |
+
"eval_steps_per_second": 9.11,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 1.07,
|
| 250 |
+
"grad_norm": 3.311278820037842,
|
| 251 |
+
"learning_rate": 0.00015737024221453287,
|
| 252 |
+
"loss": 0.825,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 1.11,
|
| 257 |
+
"grad_norm": 4.0047454833984375,
|
| 258 |
+
"learning_rate": 0.00015598615916955017,
|
| 259 |
+
"loss": 0.7347,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 1.14,
|
| 264 |
+
"grad_norm": 4.145818710327148,
|
| 265 |
+
"learning_rate": 0.00015460207612456747,
|
| 266 |
+
"loss": 0.8906,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 1.18,
|
| 271 |
+
"grad_norm": 5.543643951416016,
|
| 272 |
+
"learning_rate": 0.00015321799307958477,
|
| 273 |
+
"loss": 0.7669,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 1.21,
|
| 278 |
+
"grad_norm": 5.490930080413818,
|
| 279 |
+
"learning_rate": 0.00015183391003460207,
|
| 280 |
+
"loss": 0.774,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 1.25,
|
| 285 |
+
"grad_norm": 4.895139217376709,
|
| 286 |
+
"learning_rate": 0.00015044982698961937,
|
| 287 |
+
"loss": 0.6658,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 1.28,
|
| 292 |
+
"grad_norm": 4.527533054351807,
|
| 293 |
+
"learning_rate": 0.00014906574394463667,
|
| 294 |
+
"loss": 0.8603,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
+
"epoch": 1.31,
|
| 299 |
+
"grad_norm": 3.3609440326690674,
|
| 300 |
+
"learning_rate": 0.00014768166089965397,
|
| 301 |
+
"loss": 0.7099,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 1.35,
|
| 306 |
+
"grad_norm": 10.356605529785156,
|
| 307 |
+
"learning_rate": 0.0001462975778546713,
|
| 308 |
+
"loss": 0.7262,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 1.38,
|
| 313 |
+
"grad_norm": 3.6397433280944824,
|
| 314 |
+
"learning_rate": 0.0001449134948096886,
|
| 315 |
+
"loss": 0.7307,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
+
"epoch": 1.38,
|
| 320 |
+
"eval_accuracy": 0.8319018404907975,
|
| 321 |
+
"eval_loss": 0.577174186706543,
|
| 322 |
+
"eval_runtime": 11.1343,
|
| 323 |
+
"eval_samples_per_second": 73.197,
|
| 324 |
+
"eval_steps_per_second": 9.161,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 1.42,
|
| 329 |
+
"grad_norm": 5.693297386169434,
|
| 330 |
+
"learning_rate": 0.0001435294117647059,
|
| 331 |
+
"loss": 0.8061,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 1.45,
|
| 336 |
+
"grad_norm": 3.4219260215759277,
|
| 337 |
+
"learning_rate": 0.0001421453287197232,
|
| 338 |
+
"loss": 0.6227,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 1.49,
|
| 343 |
+
"grad_norm": 9.225582122802734,
|
| 344 |
+
"learning_rate": 0.0001407612456747405,
|
| 345 |
+
"loss": 0.6429,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
+
"epoch": 1.52,
|
| 350 |
+
"grad_norm": 5.8882293701171875,
|
| 351 |
+
"learning_rate": 0.0001393771626297578,
|
| 352 |
+
"loss": 0.686,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 1.56,
|
| 357 |
+
"grad_norm": 3.3229711055755615,
|
| 358 |
+
"learning_rate": 0.0001379930795847751,
|
| 359 |
+
"loss": 0.7574,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 1.59,
|
| 364 |
+
"grad_norm": 5.165830135345459,
|
| 365 |
+
"learning_rate": 0.0001366089965397924,
|
| 366 |
+
"loss": 0.6195,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 1.63,
|
| 371 |
+
"grad_norm": 6.676501274108887,
|
| 372 |
+
"learning_rate": 0.0001352249134948097,
|
| 373 |
+
"loss": 0.6789,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 1.66,
|
| 378 |
+
"grad_norm": 6.1190643310546875,
|
| 379 |
+
"learning_rate": 0.000133840830449827,
|
| 380 |
+
"loss": 0.7149,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 1.7,
|
| 385 |
+
"grad_norm": 2.723904848098755,
|
| 386 |
+
"learning_rate": 0.0001324567474048443,
|
| 387 |
+
"loss": 0.7014,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
+
"epoch": 1.73,
|
| 392 |
+
"grad_norm": 4.995339870452881,
|
| 393 |
+
"learning_rate": 0.00013107266435986162,
|
| 394 |
+
"loss": 0.6624,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 1.73,
|
| 399 |
+
"eval_accuracy": 0.7717791411042945,
|
| 400 |
+
"eval_loss": 0.6924724578857422,
|
| 401 |
+
"eval_runtime": 11.3854,
|
| 402 |
+
"eval_samples_per_second": 71.583,
|
| 403 |
+
"eval_steps_per_second": 8.959,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 1.76,
|
| 408 |
+
"grad_norm": 5.440706729888916,
|
| 409 |
+
"learning_rate": 0.00012968858131487892,
|
| 410 |
+
"loss": 0.8681,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 1.8,
|
| 415 |
+
"grad_norm": 4.539951801300049,
|
| 416 |
+
"learning_rate": 0.0001283044982698962,
|
| 417 |
+
"loss": 0.5997,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 1.83,
|
| 422 |
+
"grad_norm": 4.5558013916015625,
|
| 423 |
+
"learning_rate": 0.0001269204152249135,
|
| 424 |
+
"loss": 0.6839,
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 1.87,
|
| 429 |
+
"grad_norm": 4.361762046813965,
|
| 430 |
+
"learning_rate": 0.0001255363321799308,
|
| 431 |
+
"loss": 0.6016,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
+
"epoch": 1.9,
|
| 436 |
+
"grad_norm": 4.278672695159912,
|
| 437 |
+
"learning_rate": 0.0001241522491349481,
|
| 438 |
+
"loss": 0.6111,
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
+
"epoch": 1.94,
|
| 443 |
+
"grad_norm": 5.990556240081787,
|
| 444 |
+
"learning_rate": 0.0001227681660899654,
|
| 445 |
+
"loss": 0.9729,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 1.97,
|
| 450 |
+
"grad_norm": 4.511960983276367,
|
| 451 |
+
"learning_rate": 0.0001213840830449827,
|
| 452 |
+
"loss": 0.6829,
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 2.01,
|
| 457 |
+
"grad_norm": 3.139665365219116,
|
| 458 |
+
"learning_rate": 0.00012,
|
| 459 |
+
"loss": 0.6516,
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"epoch": 2.04,
|
| 464 |
+
"grad_norm": 2.4773037433624268,
|
| 465 |
+
"learning_rate": 0.0001186159169550173,
|
| 466 |
+
"loss": 0.492,
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 2.08,
|
| 471 |
+
"grad_norm": 1.613572597503662,
|
| 472 |
+
"learning_rate": 0.0001172318339100346,
|
| 473 |
+
"loss": 0.5781,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 2.08,
|
| 478 |
+
"eval_accuracy": 0.8404907975460123,
|
| 479 |
+
"eval_loss": 0.5438538789749146,
|
| 480 |
+
"eval_runtime": 11.1336,
|
| 481 |
+
"eval_samples_per_second": 73.202,
|
| 482 |
+
"eval_steps_per_second": 9.161,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 2.11,
|
| 487 |
+
"grad_norm": 2.5122666358947754,
|
| 488 |
+
"learning_rate": 0.0001158477508650519,
|
| 489 |
+
"loss": 0.55,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"epoch": 2.15,
|
| 494 |
+
"grad_norm": 5.037026882171631,
|
| 495 |
+
"learning_rate": 0.00011446366782006921,
|
| 496 |
+
"loss": 0.5404,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 2.18,
|
| 501 |
+
"grad_norm": 6.795055866241455,
|
| 502 |
+
"learning_rate": 0.00011307958477508651,
|
| 503 |
+
"loss": 0.6455,
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 2.21,
|
| 508 |
+
"grad_norm": 6.474560260772705,
|
| 509 |
+
"learning_rate": 0.00011169550173010381,
|
| 510 |
+
"loss": 0.5184,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
+
"epoch": 2.25,
|
| 515 |
+
"grad_norm": 8.110054969787598,
|
| 516 |
+
"learning_rate": 0.00011031141868512111,
|
| 517 |
+
"loss": 0.4706,
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 2.28,
|
| 522 |
+
"grad_norm": 2.1118876934051514,
|
| 523 |
+
"learning_rate": 0.00010892733564013841,
|
| 524 |
+
"loss": 0.5291,
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
+
"epoch": 2.32,
|
| 529 |
+
"grad_norm": 5.386865615844727,
|
| 530 |
+
"learning_rate": 0.00010754325259515571,
|
| 531 |
+
"loss": 0.605,
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
+
"epoch": 2.35,
|
| 536 |
+
"grad_norm": 4.501631736755371,
|
| 537 |
+
"learning_rate": 0.00010615916955017303,
|
| 538 |
+
"loss": 0.4696,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 2.39,
|
| 543 |
+
"grad_norm": 4.06800651550293,
|
| 544 |
+
"learning_rate": 0.00010477508650519033,
|
| 545 |
+
"loss": 0.5214,
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 2.42,
|
| 550 |
+
"grad_norm": 3.3574037551879883,
|
| 551 |
+
"learning_rate": 0.00010339100346020762,
|
| 552 |
+
"loss": 0.5537,
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"epoch": 2.42,
|
| 557 |
+
"eval_accuracy": 0.8331288343558282,
|
| 558 |
+
"eval_loss": 0.5256926417350769,
|
| 559 |
+
"eval_runtime": 11.3468,
|
| 560 |
+
"eval_samples_per_second": 71.826,
|
| 561 |
+
"eval_steps_per_second": 8.989,
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 2.46,
|
| 566 |
+
"grad_norm": 6.940188884735107,
|
| 567 |
+
"learning_rate": 0.00010200692041522492,
|
| 568 |
+
"loss": 0.5371,
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 2.49,
|
| 573 |
+
"grad_norm": 5.517166614532471,
|
| 574 |
+
"learning_rate": 0.00010076124567474047,
|
| 575 |
+
"loss": 0.5419,
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 2.53,
|
| 580 |
+
"grad_norm": 3.2714200019836426,
|
| 581 |
+
"learning_rate": 9.937716262975779e-05,
|
| 582 |
+
"loss": 0.6657,
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 2.56,
|
| 587 |
+
"grad_norm": 2.734272003173828,
|
| 588 |
+
"learning_rate": 9.79930795847751e-05,
|
| 589 |
+
"loss": 0.5712,
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 2.6,
|
| 594 |
+
"grad_norm": 7.275644302368164,
|
| 595 |
+
"learning_rate": 9.66089965397924e-05,
|
| 596 |
+
"loss": 0.6513,
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 2.63,
|
| 601 |
+
"grad_norm": 4.910625457763672,
|
| 602 |
+
"learning_rate": 9.52249134948097e-05,
|
| 603 |
+
"loss": 0.5641,
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 2.66,
|
| 608 |
+
"grad_norm": 2.4771077632904053,
|
| 609 |
+
"learning_rate": 9.384083044982698e-05,
|
| 610 |
+
"loss": 0.4163,
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 2.7,
|
| 615 |
+
"grad_norm": 4.3278303146362305,
|
| 616 |
+
"learning_rate": 9.24567474048443e-05,
|
| 617 |
+
"loss": 0.5759,
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 2.73,
|
| 622 |
+
"grad_norm": 8.122814178466797,
|
| 623 |
+
"learning_rate": 9.10726643598616e-05,
|
| 624 |
+
"loss": 0.6389,
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 2.77,
|
| 629 |
+
"grad_norm": 6.369782447814941,
|
| 630 |
+
"learning_rate": 8.96885813148789e-05,
|
| 631 |
+
"loss": 0.4112,
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 2.77,
|
| 636 |
+
"eval_accuracy": 0.8564417177914111,
|
| 637 |
+
"eval_loss": 0.4499729573726654,
|
| 638 |
+
"eval_runtime": 11.5138,
|
| 639 |
+
"eval_samples_per_second": 70.785,
|
| 640 |
+
"eval_steps_per_second": 8.859,
|
| 641 |
"step": 800
|
| 642 |
},
|
| 643 |
{
|
| 644 |
+
"epoch": 2.8,
|
| 645 |
+
"grad_norm": 7.130208969116211,
|
| 646 |
+
"learning_rate": 8.83044982698962e-05,
|
| 647 |
+
"loss": 0.4534,
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
+
"epoch": 2.84,
|
| 652 |
+
"grad_norm": 5.094985008239746,
|
| 653 |
+
"learning_rate": 8.69204152249135e-05,
|
| 654 |
+
"loss": 0.4881,
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
+
"epoch": 2.87,
|
| 659 |
+
"grad_norm": 2.527794361114502,
|
| 660 |
+
"learning_rate": 8.55363321799308e-05,
|
| 661 |
+
"loss": 0.2775,
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 2.91,
|
| 666 |
+
"grad_norm": 9.366188049316406,
|
| 667 |
+
"learning_rate": 8.415224913494811e-05,
|
| 668 |
+
"loss": 0.6528,
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
+
"epoch": 2.94,
|
| 673 |
+
"grad_norm": 3.863041400909424,
|
| 674 |
+
"learning_rate": 8.276816608996541e-05,
|
| 675 |
+
"loss": 0.5632,
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
+
"epoch": 2.98,
|
| 680 |
+
"grad_norm": 3.3894691467285156,
|
| 681 |
+
"learning_rate": 8.138408304498271e-05,
|
| 682 |
+
"loss": 0.5885,
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
+
"epoch": 3.01,
|
| 687 |
+
"grad_norm": 5.836617469787598,
|
| 688 |
+
"learning_rate": 8e-05,
|
| 689 |
+
"loss": 0.5864,
|
| 690 |
"step": 870
|
| 691 |
},
|
| 692 |
{
|
| 693 |
+
"epoch": 3.04,
|
| 694 |
+
"grad_norm": 4.380435943603516,
|
| 695 |
+
"learning_rate": 7.86159169550173e-05,
|
| 696 |
+
"loss": 0.4523,
|
| 697 |
"step": 880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
+
"epoch": 3.08,
|
| 701 |
+
"grad_norm": 0.4667840898036957,
|
| 702 |
+
"learning_rate": 7.72318339100346e-05,
|
| 703 |
+
"loss": 0.2999,
|
| 704 |
"step": 890
|
| 705 |
},
|
| 706 |
{
|
| 707 |
+
"epoch": 3.11,
|
| 708 |
+
"grad_norm": 3.003122091293335,
|
| 709 |
+
"learning_rate": 7.58477508650519e-05,
|
| 710 |
+
"loss": 0.3263,
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"epoch": 3.11,
|
| 715 |
+
"eval_accuracy": 0.841717791411043,
|
| 716 |
+
"eval_loss": 0.4910809397697449,
|
| 717 |
+
"eval_runtime": 11.4121,
|
| 718 |
+
"eval_samples_per_second": 71.416,
|
| 719 |
+
"eval_steps_per_second": 8.938,
|
| 720 |
"step": 900
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 3.15,
|
| 724 |
+
"grad_norm": 4.519166469573975,
|
| 725 |
+
"learning_rate": 7.44636678200692e-05,
|
| 726 |
+
"loss": 0.417,
|
| 727 |
"step": 910
|
| 728 |
},
|
| 729 |
{
|
| 730 |
+
"epoch": 3.18,
|
| 731 |
+
"grad_norm": 6.071402549743652,
|
| 732 |
+
"learning_rate": 7.30795847750865e-05,
|
| 733 |
+
"loss": 0.351,
|
| 734 |
"step": 920
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 3.22,
|
| 738 |
+
"grad_norm": 6.0896687507629395,
|
| 739 |
+
"learning_rate": 7.16955017301038e-05,
|
| 740 |
+
"loss": 0.4687,
|
| 741 |
"step": 930
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 3.25,
|
| 745 |
+
"grad_norm": 4.001879692077637,
|
| 746 |
+
"learning_rate": 7.031141868512112e-05,
|
| 747 |
+
"loss": 0.5083,
|
| 748 |
"step": 940
|
| 749 |
},
|
| 750 |
{
|
| 751 |
+
"epoch": 3.29,
|
| 752 |
+
"grad_norm": 5.064093112945557,
|
| 753 |
+
"learning_rate": 6.892733564013842e-05,
|
| 754 |
+
"loss": 0.3407,
|
| 755 |
"step": 950
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 3.32,
|
| 759 |
+
"grad_norm": 1.3715455532073975,
|
| 760 |
+
"learning_rate": 6.754325259515572e-05,
|
| 761 |
+
"loss": 0.4073,
|
| 762 |
"step": 960
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 3.36,
|
| 766 |
+
"grad_norm": 6.380220413208008,
|
| 767 |
+
"learning_rate": 6.615916955017302e-05,
|
| 768 |
+
"loss": 0.4613,
|
| 769 |
"step": 970
|
| 770 |
},
|
| 771 |
{
|
| 772 |
+
"epoch": 3.39,
|
| 773 |
+
"grad_norm": 2.1010074615478516,
|
| 774 |
+
"learning_rate": 6.477508650519032e-05,
|
| 775 |
+
"loss": 0.3706,
|
| 776 |
"step": 980
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 3.43,
|
| 780 |
+
"grad_norm": 2.2640676498413086,
|
| 781 |
+
"learning_rate": 6.339100346020761e-05,
|
| 782 |
+
"loss": 0.3235,
|
| 783 |
"step": 990
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"epoch": 3.46,
|
| 787 |
+
"grad_norm": 2.6152210235595703,
|
| 788 |
+
"learning_rate": 6.200692041522491e-05,
|
| 789 |
+
"loss": 0.4592,
|
| 790 |
"step": 1000
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"epoch": 3.46,
|
| 794 |
+
"eval_accuracy": 0.8711656441717791,
|
| 795 |
+
"eval_loss": 0.4550653100013733,
|
| 796 |
+
"eval_runtime": 11.3924,
|
| 797 |
+
"eval_samples_per_second": 71.539,
|
| 798 |
+
"eval_steps_per_second": 8.953,
|
| 799 |
"step": 1000
|
| 800 |
},
|
| 801 |
{
|
| 802 |
+
"epoch": 3.49,
|
| 803 |
+
"grad_norm": 4.868244647979736,
|
| 804 |
+
"learning_rate": 6.0622837370242214e-05,
|
| 805 |
+
"loss": 0.3891,
|
| 806 |
"step": 1010
|
| 807 |
},
|
| 808 |
{
|
| 809 |
+
"epoch": 3.53,
|
| 810 |
+
"grad_norm": 4.125463962554932,
|
| 811 |
+
"learning_rate": 5.9238754325259514e-05,
|
| 812 |
+
"loss": 0.3462,
|
| 813 |
"step": 1020
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"epoch": 3.56,
|
| 817 |
+
"grad_norm": 5.510716915130615,
|
| 818 |
+
"learning_rate": 5.785467128027682e-05,
|
| 819 |
+
"loss": 0.3938,
|
| 820 |
"step": 1030
|
| 821 |
},
|
| 822 |
{
|
| 823 |
+
"epoch": 3.6,
|
| 824 |
+
"grad_norm": 4.867416858673096,
|
| 825 |
+
"learning_rate": 5.647058823529412e-05,
|
| 826 |
+
"loss": 0.3851,
|
| 827 |
"step": 1040
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 3.63,
|
| 831 |
+
"grad_norm": 5.599556922912598,
|
| 832 |
+
"learning_rate": 5.508650519031142e-05,
|
| 833 |
+
"loss": 0.3939,
|
| 834 |
"step": 1050
|
| 835 |
},
|
| 836 |
{
|
| 837 |
+
"epoch": 3.67,
|
| 838 |
+
"grad_norm": 6.144674777984619,
|
| 839 |
+
"learning_rate": 5.3702422145328725e-05,
|
| 840 |
+
"loss": 0.5787,
|
| 841 |
"step": 1060
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"epoch": 3.7,
|
| 845 |
+
"grad_norm": 6.78063440322876,
|
| 846 |
+
"learning_rate": 5.2318339100346025e-05,
|
| 847 |
+
"loss": 0.4548,
|
| 848 |
"step": 1070
|
| 849 |
},
|
| 850 |
{
|
| 851 |
+
"epoch": 3.74,
|
| 852 |
+
"grad_norm": 2.213515281677246,
|
| 853 |
+
"learning_rate": 5.0934256055363325e-05,
|
| 854 |
+
"loss": 0.418,
|
| 855 |
"step": 1080
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 3.77,
|
| 859 |
+
"grad_norm": 3.3876266479492188,
|
| 860 |
+
"learning_rate": 4.9550173010380624e-05,
|
| 861 |
+
"loss": 0.432,
|
| 862 |
"step": 1090
|
| 863 |
},
|
| 864 |
{
|
| 865 |
+
"epoch": 3.81,
|
| 866 |
+
"grad_norm": 6.500645160675049,
|
| 867 |
+
"learning_rate": 4.8166089965397924e-05,
|
| 868 |
+
"loss": 0.3204,
|
| 869 |
"step": 1100
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 3.81,
|
| 873 |
+
"eval_accuracy": 0.8723926380368098,
|
| 874 |
+
"eval_loss": 0.4324829876422882,
|
| 875 |
+
"eval_runtime": 11.2965,
|
| 876 |
+
"eval_samples_per_second": 72.146,
|
| 877 |
+
"eval_steps_per_second": 9.029,
|
| 878 |
"step": 1100
|
| 879 |
},
|
| 880 |
{
|
| 881 |
+
"epoch": 3.84,
|
| 882 |
+
"grad_norm": 2.518057107925415,
|
| 883 |
+
"learning_rate": 4.678200692041523e-05,
|
| 884 |
+
"loss": 0.3505,
|
| 885 |
"step": 1110
|
| 886 |
},
|
| 887 |
{
|
| 888 |
+
"epoch": 3.88,
|
| 889 |
+
"grad_norm": 4.820311546325684,
|
| 890 |
+
"learning_rate": 4.539792387543253e-05,
|
| 891 |
+
"loss": 0.4641,
|
| 892 |
"step": 1120
|
| 893 |
},
|
| 894 |
{
|
| 895 |
+
"epoch": 3.91,
|
| 896 |
+
"grad_norm": 1.1859557628631592,
|
| 897 |
+
"learning_rate": 4.401384083044983e-05,
|
| 898 |
+
"loss": 0.5202,
|
| 899 |
"step": 1130
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 3.94,
|
| 903 |
+
"grad_norm": 2.3869946002960205,
|
| 904 |
+
"learning_rate": 4.262975778546713e-05,
|
| 905 |
+
"loss": 0.5007,
|
| 906 |
"step": 1140
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 3.98,
|
| 910 |
+
"grad_norm": 6.64663553237915,
|
| 911 |
+
"learning_rate": 4.124567474048443e-05,
|
| 912 |
+
"loss": 0.3478,
|
| 913 |
"step": 1150
|
| 914 |
},
|
| 915 |
{
|
| 916 |
+
"epoch": 4.01,
|
| 917 |
+
"grad_norm": 4.931898593902588,
|
| 918 |
+
"learning_rate": 3.9861591695501735e-05,
|
| 919 |
+
"loss": 0.3808,
|
| 920 |
"step": 1160
|
| 921 |
},
|
| 922 |
{
|
| 923 |
+
"epoch": 4.05,
|
| 924 |
+
"grad_norm": 1.7521384954452515,
|
| 925 |
+
"learning_rate": 3.8477508650519034e-05,
|
| 926 |
+
"loss": 0.3247,
|
| 927 |
"step": 1170
|
| 928 |
},
|
| 929 |
{
|
| 930 |
+
"epoch": 4.08,
|
| 931 |
+
"grad_norm": 1.5513068437576294,
|
| 932 |
+
"learning_rate": 3.7093425605536334e-05,
|
| 933 |
+
"loss": 0.393,
|
| 934 |
"step": 1180
|
| 935 |
},
|
| 936 |
{
|
| 937 |
+
"epoch": 4.12,
|
| 938 |
+
"grad_norm": 5.2524094581604,
|
| 939 |
+
"learning_rate": 3.570934256055363e-05,
|
| 940 |
+
"loss": 0.3697,
|
| 941 |
"step": 1190
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"epoch": 4.15,
|
| 945 |
+
"grad_norm": 5.346503734588623,
|
| 946 |
+
"learning_rate": 3.432525951557093e-05,
|
| 947 |
+
"loss": 0.3268,
|
| 948 |
"step": 1200
|
| 949 |
},
|
| 950 |
{
|
| 951 |
+
"epoch": 4.15,
|
| 952 |
+
"eval_accuracy": 0.8539877300613496,
|
| 953 |
+
"eval_loss": 0.4529338777065277,
|
| 954 |
+
"eval_runtime": 11.243,
|
| 955 |
+
"eval_samples_per_second": 72.49,
|
| 956 |
+
"eval_steps_per_second": 9.072,
|
| 957 |
"step": 1200
|
| 958 |
},
|
| 959 |
{
|
| 960 |
+
"epoch": 4.19,
|
| 961 |
+
"grad_norm": 2.0786306858062744,
|
| 962 |
+
"learning_rate": 3.294117647058824e-05,
|
| 963 |
+
"loss": 0.2967,
|
| 964 |
"step": 1210
|
| 965 |
},
|
| 966 |
{
|
| 967 |
+
"epoch": 4.22,
|
| 968 |
+
"grad_norm": 3.8141846656799316,
|
| 969 |
+
"learning_rate": 3.155709342560554e-05,
|
| 970 |
+
"loss": 0.2483,
|
| 971 |
"step": 1220
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"epoch": 4.26,
|
| 975 |
+
"grad_norm": 1.8864765167236328,
|
| 976 |
+
"learning_rate": 3.0173010380622842e-05,
|
| 977 |
+
"loss": 0.2443,
|
| 978 |
"step": 1230
|
| 979 |
},
|
| 980 |
{
|
| 981 |
+
"epoch": 4.29,
|
| 982 |
+
"grad_norm": 0.9406078457832336,
|
| 983 |
+
"learning_rate": 2.878892733564014e-05,
|
| 984 |
+
"loss": 0.2637,
|
| 985 |
"step": 1240
|
| 986 |
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 4.33,
|
| 989 |
+
"grad_norm": 3.418455123901367,
|
| 990 |
+
"learning_rate": 2.7404844290657437e-05,
|
| 991 |
+
"loss": 0.2866,
|
| 992 |
+
"step": 1250
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 4.36,
|
| 996 |
+
"grad_norm": 1.8662397861480713,
|
| 997 |
+
"learning_rate": 2.602076124567474e-05,
|
| 998 |
+
"loss": 0.3381,
|
| 999 |
+
"step": 1260
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 4.39,
|
| 1003 |
+
"grad_norm": 3.9735002517700195,
|
| 1004 |
+
"learning_rate": 2.4636678200692043e-05,
|
| 1005 |
+
"loss": 0.3328,
|
| 1006 |
+
"step": 1270
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 4.43,
|
| 1010 |
+
"grad_norm": 2.229581117630005,
|
| 1011 |
+
"learning_rate": 2.3252595155709346e-05,
|
| 1012 |
+
"loss": 0.4579,
|
| 1013 |
+
"step": 1280
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 4.46,
|
| 1017 |
+
"grad_norm": 3.8186144828796387,
|
| 1018 |
+
"learning_rate": 2.1868512110726642e-05,
|
| 1019 |
+
"loss": 0.2986,
|
| 1020 |
+
"step": 1290
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 4.5,
|
| 1024 |
+
"grad_norm": 2.310910701751709,
|
| 1025 |
+
"learning_rate": 2.0484429065743945e-05,
|
| 1026 |
+
"loss": 0.4267,
|
| 1027 |
+
"step": 1300
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 4.5,
|
| 1031 |
+
"eval_accuracy": 0.8723926380368098,
|
| 1032 |
+
"eval_loss": 0.4355594515800476,
|
| 1033 |
+
"eval_runtime": 11.2323,
|
| 1034 |
+
"eval_samples_per_second": 72.559,
|
| 1035 |
+
"eval_steps_per_second": 9.081,
|
| 1036 |
+
"step": 1300
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 4.53,
|
| 1040 |
+
"grad_norm": 5.4986042976379395,
|
| 1041 |
+
"learning_rate": 1.910034602076125e-05,
|
| 1042 |
+
"loss": 0.3289,
|
| 1043 |
+
"step": 1310
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 4.57,
|
| 1047 |
+
"grad_norm": 9.17880916595459,
|
| 1048 |
+
"learning_rate": 1.7716262975778548e-05,
|
| 1049 |
+
"loss": 0.4384,
|
| 1050 |
+
"step": 1320
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 4.6,
|
| 1054 |
+
"grad_norm": 0.5825958251953125,
|
| 1055 |
+
"learning_rate": 1.6332179930795848e-05,
|
| 1056 |
+
"loss": 0.1629,
|
| 1057 |
+
"step": 1330
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 4.64,
|
| 1061 |
+
"grad_norm": 3.7861948013305664,
|
| 1062 |
+
"learning_rate": 1.4948096885813149e-05,
|
| 1063 |
+
"loss": 0.2607,
|
| 1064 |
+
"step": 1340
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 4.67,
|
| 1068 |
+
"grad_norm": 0.08592710644006729,
|
| 1069 |
+
"learning_rate": 1.356401384083045e-05,
|
| 1070 |
+
"loss": 0.23,
|
| 1071 |
+
"step": 1350
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 4.71,
|
| 1075 |
+
"grad_norm": 3.897566556930542,
|
| 1076 |
+
"learning_rate": 1.2179930795847751e-05,
|
| 1077 |
+
"loss": 0.2927,
|
| 1078 |
+
"step": 1360
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 4.74,
|
| 1082 |
+
"grad_norm": 3.4900457859039307,
|
| 1083 |
+
"learning_rate": 1.0795847750865053e-05,
|
| 1084 |
+
"loss": 0.3149,
|
| 1085 |
+
"step": 1370
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 4.78,
|
| 1089 |
+
"grad_norm": 0.48032164573669434,
|
| 1090 |
+
"learning_rate": 9.411764705882354e-06,
|
| 1091 |
+
"loss": 0.3346,
|
| 1092 |
+
"step": 1380
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 4.81,
|
| 1096 |
+
"grad_norm": 3.148484945297241,
|
| 1097 |
+
"learning_rate": 8.027681660899653e-06,
|
| 1098 |
+
"loss": 0.3006,
|
| 1099 |
+
"step": 1390
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 4.84,
|
| 1103 |
+
"grad_norm": 8.376299858093262,
|
| 1104 |
+
"learning_rate": 6.6435986159169555e-06,
|
| 1105 |
+
"loss": 0.2886,
|
| 1106 |
+
"step": 1400
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 4.84,
|
| 1110 |
+
"eval_accuracy": 0.8785276073619632,
|
| 1111 |
+
"eval_loss": 0.4145749807357788,
|
| 1112 |
+
"eval_runtime": 11.483,
|
| 1113 |
+
"eval_samples_per_second": 70.974,
|
| 1114 |
+
"eval_steps_per_second": 8.883,
|
| 1115 |
+
"step": 1400
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 4.88,
|
| 1119 |
+
"grad_norm": 10.592811584472656,
|
| 1120 |
+
"learning_rate": 5.259515570934257e-06,
|
| 1121 |
+
"loss": 0.2866,
|
| 1122 |
+
"step": 1410
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 4.91,
|
| 1126 |
+
"grad_norm": 1.9005374908447266,
|
| 1127 |
+
"learning_rate": 3.875432525951557e-06,
|
| 1128 |
+
"loss": 0.367,
|
| 1129 |
+
"step": 1420
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 4.95,
|
| 1133 |
+
"grad_norm": 6.674309730529785,
|
| 1134 |
+
"learning_rate": 2.4913494809688584e-06,
|
| 1135 |
+
"loss": 0.2715,
|
| 1136 |
+
"step": 1430
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 4.98,
|
| 1140 |
+
"grad_norm": 7.622613430023193,
|
| 1141 |
+
"learning_rate": 1.1072664359861592e-06,
|
| 1142 |
+
"loss": 0.3306,
|
| 1143 |
+
"step": 1440
|
| 1144 |
+
},
|
| 1145 |
{
|
| 1146 |
"epoch": 5.0,
|
| 1147 |
+
"step": 1445,
|
| 1148 |
+
"total_flos": 1.789030847196795e+18,
|
| 1149 |
+
"train_loss": 0.6435174308433664,
|
| 1150 |
+
"train_runtime": 776.1745,
|
| 1151 |
+
"train_samples_per_second": 29.742,
|
| 1152 |
+
"train_steps_per_second": 1.862
|
| 1153 |
}
|
| 1154 |
],
|
| 1155 |
"logging_steps": 10,
|
| 1156 |
+
"max_steps": 1445,
|
| 1157 |
+
"num_input_tokens_seen": 0,
|
| 1158 |
"num_train_epochs": 5,
|
| 1159 |
"save_steps": 100,
|
| 1160 |
+
"total_flos": 1.789030847196795e+18,
|
| 1161 |
+
"train_batch_size": 16,
|
| 1162 |
"trial_name": null,
|
| 1163 |
"trial_params": null
|
| 1164 |
}
|