Commit
•
0e9a7b6
1
Parent(s):
c0e2beb
End of training
Browse files- README.md +3 -1
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/May06_10-03-49_nmjti7f45r/events.out.tfevents.1714995748.nmjti7f45r.226.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +1591 -549
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/convnextv2-tiny-1k-224
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -15,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# convnextv2-tiny-1k-224-finetuned-galaxy10-decals
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [facebook/convnextv2-tiny-1k-224](https://huggingface.co/facebook/convnextv2-tiny-1k-224) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.4261
|
| 21 |
- Accuracy: 0.8703
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/convnextv2-tiny-1k-224
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
+
- vision
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
|
|
|
| 17 |
|
| 18 |
# convnextv2-tiny-1k-224-finetuned-galaxy10-decals
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [facebook/convnextv2-tiny-1k-224](https://huggingface.co/facebook/convnextv2-tiny-1k-224) on the matthieulel/galaxy10_decals dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
- Loss: 0.4261
|
| 23 |
- Accuracy: 0.8703
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 19.879759519038075,
|
| 3 |
+
"eval_accuracy": 0.8703494926719278,
|
| 4 |
+
"eval_loss": 0.4261245131492615,
|
| 5 |
+
"eval_runtime": 25.4134,
|
| 6 |
+
"eval_samples_per_second": 69.806,
|
| 7 |
+
"eval_steps_per_second": 2.204,
|
| 8 |
+
"total_flos": 7.988705158075343e+18,
|
| 9 |
+
"train_loss": 0.5653726263392356,
|
| 10 |
+
"train_runtime": 5886.8979,
|
| 11 |
+
"train_samples_per_second": 54.229,
|
| 12 |
+
"train_steps_per_second": 0.421
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 19.879759519038075,
|
| 3 |
+
"eval_accuracy": 0.8703494926719278,
|
| 4 |
+
"eval_loss": 0.4261245131492615,
|
| 5 |
+
"eval_runtime": 25.4134,
|
| 6 |
+
"eval_samples_per_second": 69.806,
|
| 7 |
+
"eval_steps_per_second": 2.204
|
| 8 |
}
|
runs/May06_10-03-49_nmjti7f45r/events.out.tfevents.1714995748.nmjti7f45r.226.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d906b679414f6f7b01893920792ceeed5c78b6db17d357bece5b695a5dc7498
|
| 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":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 19.879759519038075,
|
| 3 |
+
"total_flos": 7.988705158075343e+18,
|
| 4 |
+
"train_loss": 0.5653726263392356,
|
| 5 |
+
"train_runtime": 5886.8979,
|
| 6 |
+
"train_samples_per_second": 54.229,
|
| 7 |
+
"train_steps_per_second": 0.421
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,903 +1,1945 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "convnextv2-tiny-1k-224-finetuned-galaxy10-decals/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.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 2.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 2.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss":
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 1.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"
|
| 91 |
-
"
|
| 92 |
-
"
|
| 93 |
-
"
|
| 94 |
-
"eval_steps_per_second": 4.071,
|
| 95 |
-
"step": 112
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch":
|
| 99 |
-
"
|
| 100 |
-
"
|
| 101 |
-
"
|
| 102 |
-
"
|
|
|
|
|
|
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 1.
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate":
|
| 108 |
-
"loss": 1.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
-
"epoch": 1.
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss":
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 1.
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss":
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 1.
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate":
|
| 129 |
-
"loss":
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"epoch": 1.
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss":
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch": 1.
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate":
|
| 143 |
-
"loss":
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch": 1.
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"learning_rate":
|
| 150 |
-
"loss":
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
-
"epoch": 1.
|
| 155 |
-
"grad_norm":
|
| 156 |
-
"learning_rate": 4.
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch": 1.
|
| 162 |
-
"grad_norm":
|
| 163 |
-
"learning_rate": 4.
|
| 164 |
-
"loss": 0.
|
| 165 |
"step": 210
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch": 1.
|
| 169 |
-
"grad_norm":
|
| 170 |
-
"learning_rate": 4.
|
| 171 |
-
"loss": 0.
|
| 172 |
"step": 220
|
| 173 |
},
|
| 174 |
{
|
| 175 |
-
"epoch": 1.
|
| 176 |
-
"
|
| 177 |
-
"
|
| 178 |
-
"
|
| 179 |
-
"eval_samples_per_second": 129.8,
|
| 180 |
-
"eval_steps_per_second": 4.064,
|
| 181 |
-
"step": 224
|
| 182 |
-
},
|
| 183 |
-
{
|
| 184 |
-
"epoch": 2.048997772828508,
|
| 185 |
-
"grad_norm": 10.136615753173828,
|
| 186 |
-
"learning_rate": 4.41468253968254e-05,
|
| 187 |
-
"loss": 0.7221,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate": 4.
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch":
|
| 199 |
-
"
|
| 200 |
-
"
|
| 201 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch": 2.
|
| 206 |
-
"grad_norm":
|
| 207 |
-
"learning_rate": 4.
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch": 2.
|
| 213 |
-
"grad_norm":
|
| 214 |
-
"learning_rate": 4.
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch": 2.
|
| 220 |
-
"grad_norm":
|
| 221 |
-
"learning_rate": 4.
|
| 222 |
-
"loss": 0.
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
-
"epoch": 2.
|
| 227 |
-
"grad_norm":
|
| 228 |
-
"learning_rate": 4.
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch": 2.
|
| 234 |
-
"grad_norm":
|
| 235 |
-
"learning_rate": 4.
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch": 2.
|
| 241 |
-
"grad_norm":
|
| 242 |
-
"learning_rate": 4.
|
| 243 |
-
"loss": 0.
|
| 244 |
"step": 310
|
| 245 |
},
|
| 246 |
{
|
| 247 |
-
"epoch": 2.
|
| 248 |
-
"grad_norm": 13.
|
| 249 |
-
"learning_rate":
|
| 250 |
-
"loss": 0.
|
| 251 |
"step": 320
|
| 252 |
},
|
| 253 |
{
|
| 254 |
-
"epoch": 2.
|
| 255 |
-
"grad_norm":
|
| 256 |
-
"learning_rate":
|
| 257 |
-
"loss": 0.
|
| 258 |
"step": 330
|
| 259 |
},
|
| 260 |
{
|
| 261 |
-
"epoch": 2.
|
| 262 |
-
"
|
| 263 |
-
"
|
| 264 |
-
"
|
| 265 |
-
"eval_samples_per_second": 129.961,
|
| 266 |
-
"eval_steps_per_second": 4.069,
|
| 267 |
-
"step": 336
|
| 268 |
-
},
|
| 269 |
-
{
|
| 270 |
-
"epoch": 3.0289532293986636,
|
| 271 |
-
"grad_norm": 16.003482818603516,
|
| 272 |
-
"learning_rate": 3.8690476190476195e-05,
|
| 273 |
-
"loss": 0.6928,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
-
"epoch":
|
| 278 |
-
"grad_norm":
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"grad_norm":
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
-
"epoch":
|
| 292 |
-
"grad_norm":
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
-
"epoch":
|
| 299 |
-
"
|
| 300 |
-
"
|
| 301 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch": 3.
|
| 306 |
-
"grad_norm":
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch": 3.
|
| 313 |
-
"grad_norm":
|
| 314 |
-
"learning_rate":
|
| 315 |
-
"loss": 0.
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
-
"epoch": 3.
|
| 320 |
-
"grad_norm":
|
| 321 |
-
"learning_rate":
|
| 322 |
-
"loss": 0.
|
| 323 |
"step": 410
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch": 3.
|
| 327 |
-
"grad_norm":
|
| 328 |
-
"learning_rate":
|
| 329 |
-
"loss": 0.
|
| 330 |
"step": 420
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch": 3.
|
| 334 |
-
"grad_norm":
|
| 335 |
-
"learning_rate":
|
| 336 |
-
"loss": 0.
|
| 337 |
"step": 430
|
| 338 |
},
|
| 339 |
{
|
| 340 |
-
"epoch": 3.
|
| 341 |
-
"grad_norm": 12.
|
| 342 |
-
"learning_rate":
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 440
|
| 345 |
},
|
| 346 |
{
|
| 347 |
-
"epoch":
|
| 348 |
-
"
|
| 349 |
-
"
|
| 350 |
-
"
|
| 351 |
-
"eval_samples_per_second": 127.835,
|
| 352 |
-
"eval_steps_per_second": 4.002,
|
| 353 |
-
"step": 449
|
| 354 |
-
},
|
| 355 |
-
{
|
| 356 |
-
"epoch": 4.008908685968819,
|
| 357 |
-
"grad_norm": 11.317361831665039,
|
| 358 |
-
"learning_rate": 3.3234126984126983e-05,
|
| 359 |
-
"loss": 0.613,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
-
"grad_norm":
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch":
|
| 371 |
-
"grad_norm":
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch":
|
| 378 |
-
"grad_norm":
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch": 4.
|
| 392 |
-
"
|
| 393 |
-
"
|
| 394 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch": 4.
|
| 399 |
-
"grad_norm":
|
| 400 |
-
"learning_rate":
|
| 401 |
-
"loss": 0.
|
| 402 |
"step": 510
|
| 403 |
},
|
| 404 |
{
|
| 405 |
-
"epoch": 4.
|
| 406 |
-
"grad_norm": 9.
|
| 407 |
-
"learning_rate":
|
| 408 |
-
"loss": 0.
|
| 409 |
"step": 520
|
| 410 |
},
|
| 411 |
{
|
| 412 |
-
"epoch": 4.
|
| 413 |
-
"grad_norm":
|
| 414 |
-
"learning_rate":
|
| 415 |
-
"loss": 0.
|
| 416 |
"step": 530
|
| 417 |
},
|
| 418 |
{
|
| 419 |
-
"epoch": 4.
|
| 420 |
-
"grad_norm":
|
| 421 |
-
"learning_rate":
|
| 422 |
-
"loss": 0.
|
| 423 |
"step": 540
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch": 4.
|
| 427 |
-
"grad_norm": 11.
|
| 428 |
-
"learning_rate":
|
| 429 |
-
"loss": 0.
|
| 430 |
"step": 550
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"epoch": 4.
|
| 434 |
-
"grad_norm":
|
| 435 |
-
"learning_rate":
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 560
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch": 4.
|
| 441 |
-
"
|
| 442 |
-
"
|
| 443 |
-
"
|
| 444 |
-
"eval_samples_per_second": 129.93,
|
| 445 |
-
"eval_steps_per_second": 4.068,
|
| 446 |
-
"step": 561
|
| 447 |
-
},
|
| 448 |
-
{
|
| 449 |
-
"epoch": 5.077951002227172,
|
| 450 |
-
"grad_norm": 15.332996368408203,
|
| 451 |
-
"learning_rate": 2.7281746031746032e-05,
|
| 452 |
-
"loss": 0.5617,
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch":
|
| 457 |
-
"grad_norm":
|
| 458 |
-
"learning_rate":
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
-
"epoch":
|
| 464 |
-
"grad_norm": 13.
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"grad_norm":
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
-
"epoch":
|
| 478 |
-
"grad_norm":
|
| 479 |
-
"learning_rate":
|
| 480 |
-
"loss": 0.
|
| 481 |
"step": 610
|
| 482 |
},
|
| 483 |
{
|
| 484 |
-
"epoch":
|
| 485 |
-
"grad_norm":
|
| 486 |
-
"learning_rate":
|
| 487 |
-
"loss": 0.
|
| 488 |
"step": 620
|
| 489 |
},
|
| 490 |
{
|
| 491 |
-
"epoch":
|
| 492 |
-
"
|
| 493 |
-
"
|
| 494 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 495 |
"step": 630
|
| 496 |
},
|
| 497 |
{
|
| 498 |
-
"epoch": 5.
|
| 499 |
-
"grad_norm": 13.
|
| 500 |
-
"learning_rate":
|
| 501 |
-
"loss": 0.
|
| 502 |
"step": 640
|
| 503 |
},
|
| 504 |
{
|
| 505 |
-
"epoch": 5.
|
| 506 |
-
"grad_norm":
|
| 507 |
-
"learning_rate":
|
| 508 |
-
"loss": 0.
|
| 509 |
"step": 650
|
| 510 |
},
|
| 511 |
{
|
| 512 |
-
"epoch": 5.
|
| 513 |
-
"grad_norm":
|
| 514 |
-
"learning_rate":
|
| 515 |
-
"loss": 0.
|
| 516 |
"step": 660
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch": 5.
|
| 520 |
-
"grad_norm":
|
| 521 |
-
"learning_rate":
|
| 522 |
-
"loss": 0.
|
| 523 |
"step": 670
|
| 524 |
},
|
| 525 |
{
|
| 526 |
-
"epoch": 5.
|
| 527 |
-
"
|
| 528 |
-
"
|
| 529 |
-
"
|
| 530 |
-
"eval_samples_per_second": 129.914,
|
| 531 |
-
"eval_steps_per_second": 4.067,
|
| 532 |
-
"step": 673
|
| 533 |
-
},
|
| 534 |
-
{
|
| 535 |
-
"epoch": 6.057906458797327,
|
| 536 |
-
"grad_norm": 10.681763648986816,
|
| 537 |
-
"learning_rate": 2.1825396825396827e-05,
|
| 538 |
-
"loss": 0.5332,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
-
"epoch":
|
| 543 |
-
"grad_norm":
|
| 544 |
-
"learning_rate":
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
-
"epoch":
|
| 550 |
-
"grad_norm": 16.
|
| 551 |
-
"learning_rate":
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
-
"epoch":
|
| 557 |
-
"grad_norm":
|
| 558 |
-
"learning_rate":
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 710
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"epoch":
|
| 564 |
-
"grad_norm":
|
| 565 |
-
"learning_rate":
|
| 566 |
-
"loss": 0.
|
| 567 |
"step": 720
|
| 568 |
},
|
| 569 |
{
|
| 570 |
-
"epoch":
|
| 571 |
-
"grad_norm":
|
| 572 |
-
"learning_rate":
|
| 573 |
-
"loss": 0.
|
| 574 |
"step": 730
|
| 575 |
},
|
| 576 |
{
|
| 577 |
-
"epoch":
|
| 578 |
-
"grad_norm":
|
| 579 |
-
"learning_rate":
|
| 580 |
-
"loss": 0.
|
| 581 |
"step": 740
|
| 582 |
},
|
| 583 |
{
|
| 584 |
-
"epoch":
|
| 585 |
-
"
|
| 586 |
-
"
|
| 587 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 588 |
"step": 750
|
| 589 |
},
|
| 590 |
{
|
| 591 |
-
"epoch": 6.
|
| 592 |
-
"grad_norm":
|
| 593 |
-
"learning_rate":
|
| 594 |
-
"loss": 0.
|
| 595 |
"step": 760
|
| 596 |
},
|
| 597 |
{
|
| 598 |
-
"epoch": 6.
|
| 599 |
-
"grad_norm":
|
| 600 |
-
"learning_rate":
|
| 601 |
-
"loss": 0.
|
| 602 |
"step": 770
|
| 603 |
},
|
| 604 |
{
|
| 605 |
-
"epoch": 6.
|
| 606 |
-
"grad_norm":
|
| 607 |
-
"learning_rate":
|
| 608 |
-
"loss": 0.
|
| 609 |
"step": 780
|
| 610 |
},
|
| 611 |
{
|
| 612 |
-
"epoch": 6.
|
| 613 |
-
"
|
| 614 |
-
"
|
| 615 |
-
"
|
| 616 |
-
"eval_samples_per_second": 129.131,
|
| 617 |
-
"eval_steps_per_second": 4.043,
|
| 618 |
-
"step": 785
|
| 619 |
-
},
|
| 620 |
-
{
|
| 621 |
-
"epoch": 7.0378619153674835,
|
| 622 |
-
"grad_norm": 9.479693412780762,
|
| 623 |
-
"learning_rate": 1.636904761904762e-05,
|
| 624 |
-
"loss": 0.4586,
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch":
|
| 629 |
-
"grad_norm":
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch":
|
| 636 |
-
"grad_norm":
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 810
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch":
|
| 643 |
-
"grad_norm":
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 820
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch":
|
| 650 |
-
"grad_norm":
|
| 651 |
-
"learning_rate":
|
| 652 |
-
"loss": 0.
|
| 653 |
"step": 830
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch":
|
| 657 |
-
"grad_norm":
|
| 658 |
-
"learning_rate":
|
| 659 |
-
"loss": 0.
|
| 660 |
"step": 840
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch":
|
| 664 |
-
"grad_norm":
|
| 665 |
-
"learning_rate":
|
| 666 |
-
"loss": 0.
|
| 667 |
"step": 850
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch":
|
| 671 |
-
"grad_norm":
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 860
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch":
|
| 678 |
-
"grad_norm": 11.
|
| 679 |
-
"learning_rate":
|
| 680 |
-
"loss": 0.
|
| 681 |
"step": 870
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch":
|
| 685 |
-
"
|
| 686 |
-
"
|
| 687 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 688 |
"step": 880
|
| 689 |
},
|
| 690 |
{
|
| 691 |
-
"epoch": 7.
|
| 692 |
-
"grad_norm":
|
| 693 |
-
"learning_rate":
|
| 694 |
-
"loss": 0.
|
| 695 |
"step": 890
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch":
|
| 699 |
-
"
|
| 700 |
-
"
|
| 701 |
-
"
|
| 702 |
-
"eval_samples_per_second": 128.892,
|
| 703 |
-
"eval_steps_per_second": 4.035,
|
| 704 |
-
"step": 898
|
| 705 |
-
},
|
| 706 |
-
{
|
| 707 |
-
"epoch": 8.017817371937639,
|
| 708 |
-
"grad_norm": 12.601805686950684,
|
| 709 |
-
"learning_rate": 1.0912698412698414e-05,
|
| 710 |
-
"loss": 0.5064,
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"epoch":
|
| 715 |
-
"grad_norm":
|
| 716 |
-
"learning_rate":
|
| 717 |
-
"loss": 0.
|
| 718 |
"step": 910
|
| 719 |
},
|
| 720 |
{
|
| 721 |
-
"epoch":
|
| 722 |
-
"grad_norm":
|
| 723 |
-
"learning_rate":
|
| 724 |
-
"loss": 0.
|
| 725 |
"step": 920
|
| 726 |
},
|
| 727 |
{
|
| 728 |
-
"epoch":
|
| 729 |
-
"grad_norm":
|
| 730 |
-
"learning_rate":
|
| 731 |
-
"loss": 0.
|
| 732 |
"step": 930
|
| 733 |
},
|
| 734 |
{
|
| 735 |
-
"epoch":
|
| 736 |
-
"grad_norm":
|
| 737 |
-
"learning_rate":
|
| 738 |
-
"loss": 0.
|
| 739 |
"step": 940
|
| 740 |
},
|
| 741 |
{
|
| 742 |
-
"epoch":
|
| 743 |
-
"grad_norm":
|
| 744 |
-
"learning_rate":
|
| 745 |
-
"loss": 0.
|
| 746 |
"step": 950
|
| 747 |
},
|
| 748 |
{
|
| 749 |
-
"epoch":
|
| 750 |
-
"grad_norm":
|
| 751 |
-
"learning_rate":
|
| 752 |
-
"loss": 0.
|
| 753 |
"step": 960
|
| 754 |
},
|
| 755 |
{
|
| 756 |
-
"epoch":
|
| 757 |
-
"grad_norm":
|
| 758 |
-
"learning_rate":
|
| 759 |
-
"loss": 0.
|
| 760 |
"step": 970
|
| 761 |
},
|
| 762 |
{
|
| 763 |
-
"epoch":
|
| 764 |
-
"grad_norm":
|
| 765 |
-
"learning_rate":
|
| 766 |
-
"loss": 0.
|
| 767 |
"step": 980
|
| 768 |
},
|
| 769 |
{
|
| 770 |
-
"epoch":
|
| 771 |
-
"grad_norm":
|
| 772 |
-
"learning_rate":
|
| 773 |
-
"loss": 0.
|
| 774 |
"step": 990
|
| 775 |
},
|
| 776 |
{
|
| 777 |
-
"epoch": 8.
|
| 778 |
-
"
|
| 779 |
-
"
|
| 780 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 781 |
"step": 1000
|
| 782 |
},
|
| 783 |
{
|
| 784 |
-
"epoch": 8.
|
| 785 |
-
"grad_norm":
|
| 786 |
-
"learning_rate":
|
| 787 |
-
"loss": 0.
|
| 788 |
-
"step": 1010
|
| 789 |
-
},
|
| 790 |
-
{
|
| 791 |
-
"epoch": 8.997772828507795,
|
| 792 |
-
"eval_accuracy": 0.8634940513462742,
|
| 793 |
-
"eval_loss": 0.40621063113212585,
|
| 794 |
-
"eval_runtime": 12.3051,
|
| 795 |
-
"eval_samples_per_second": 129.783,
|
| 796 |
-
"eval_steps_per_second": 4.063,
|
| 797 |
"step": 1010
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch":
|
| 801 |
-
"grad_norm":
|
| 802 |
-
"learning_rate":
|
| 803 |
-
"loss": 0.
|
| 804 |
"step": 1020
|
| 805 |
},
|
| 806 |
{
|
| 807 |
-
"epoch":
|
| 808 |
-
"grad_norm":
|
| 809 |
-
"learning_rate":
|
| 810 |
-
"loss": 0.
|
| 811 |
"step": 1030
|
| 812 |
},
|
| 813 |
{
|
| 814 |
-
"epoch":
|
| 815 |
-
"grad_norm": 10.
|
| 816 |
-
"learning_rate": 3.
|
| 817 |
-
"loss": 0.
|
| 818 |
"step": 1040
|
| 819 |
},
|
| 820 |
{
|
| 821 |
-
"epoch":
|
| 822 |
-
"grad_norm":
|
| 823 |
-
"learning_rate": 3.
|
| 824 |
-
"loss": 0.
|
| 825 |
"step": 1050
|
| 826 |
},
|
| 827 |
{
|
| 828 |
-
"epoch":
|
| 829 |
-
"grad_norm":
|
| 830 |
-
"learning_rate":
|
| 831 |
-
"loss": 0.
|
| 832 |
"step": 1060
|
| 833 |
},
|
| 834 |
{
|
| 835 |
-
"epoch":
|
| 836 |
-
"grad_norm":
|
| 837 |
-
"learning_rate":
|
| 838 |
-
"loss": 0.
|
| 839 |
"step": 1070
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch":
|
| 843 |
-
"grad_norm":
|
| 844 |
-
"learning_rate":
|
| 845 |
-
"loss": 0.
|
| 846 |
"step": 1080
|
| 847 |
},
|
| 848 |
{
|
| 849 |
-
"epoch":
|
| 850 |
-
"grad_norm":
|
| 851 |
-
"learning_rate":
|
| 852 |
-
"loss": 0.
|
| 853 |
"step": 1090
|
| 854 |
},
|
| 855 |
{
|
| 856 |
-
"epoch":
|
| 857 |
-
"grad_norm":
|
| 858 |
-
"learning_rate":
|
| 859 |
-
"loss": 0.
|
| 860 |
"step": 1100
|
| 861 |
},
|
| 862 |
{
|
| 863 |
-
"epoch":
|
| 864 |
-
"grad_norm":
|
| 865 |
-
"learning_rate":
|
| 866 |
-
"loss": 0.
|
| 867 |
"step": 1110
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"epoch":
|
| 871 |
-
"grad_norm":
|
| 872 |
-
"learning_rate":
|
| 873 |
-
"loss": 0.
|
| 874 |
"step": 1120
|
| 875 |
},
|
| 876 |
{
|
| 877 |
-
"epoch":
|
| 878 |
-
"eval_accuracy": 0.
|
| 879 |
-
"eval_loss": 0.
|
| 880 |
-
"eval_runtime":
|
| 881 |
-
"eval_samples_per_second":
|
| 882 |
-
"eval_steps_per_second":
|
| 883 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 884 |
},
|
| 885 |
{
|
| 886 |
-
"epoch":
|
| 887 |
-
"
|
| 888 |
-
"
|
| 889 |
-
"
|
| 890 |
-
"
|
| 891 |
-
|
| 892 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 893 |
}
|
| 894 |
],
|
| 895 |
"logging_steps": 10,
|
| 896 |
-
"max_steps":
|
| 897 |
"num_input_tokens_seen": 0,
|
| 898 |
-
"num_train_epochs":
|
| 899 |
"save_steps": 500,
|
| 900 |
-
"total_flos":
|
| 901 |
"train_batch_size": 32,
|
| 902 |
"trial_name": null,
|
| 903 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8703494926719278,
|
| 3 |
+
"best_model_checkpoint": "convnextv2-tiny-1k-224-finetuned-galaxy10-decals/checkpoint-2480",
|
| 4 |
+
"epoch": 19.879759519038075,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 2480,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.08016032064128256,
|
| 13 |
+
"grad_norm": 5.443685531616211,
|
| 14 |
+
"learning_rate": 2.0161290322580646e-06,
|
| 15 |
+
"loss": 2.3476,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.16032064128256512,
|
| 20 |
+
"grad_norm": 5.888722896575928,
|
| 21 |
+
"learning_rate": 4.032258064516129e-06,
|
| 22 |
+
"loss": 2.3134,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.24048096192384769,
|
| 27 |
+
"grad_norm": 3.9038140773773193,
|
| 28 |
+
"learning_rate": 6.048387096774194e-06,
|
| 29 |
+
"loss": 2.2707,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.32064128256513025,
|
| 34 |
+
"grad_norm": 2.5614984035491943,
|
| 35 |
+
"learning_rate": 8.064516129032258e-06,
|
| 36 |
+
"loss": 2.2183,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.40080160320641284,
|
| 41 |
+
"grad_norm": 5.785397052764893,
|
| 42 |
+
"learning_rate": 1.0080645161290323e-05,
|
| 43 |
+
"loss": 2.1604,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.48096192384769537,
|
| 48 |
+
"grad_norm": 5.09072208404541,
|
| 49 |
+
"learning_rate": 1.2096774193548388e-05,
|
| 50 |
+
"loss": 2.0969,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.561122244488978,
|
| 55 |
+
"grad_norm": 2.9109528064727783,
|
| 56 |
+
"learning_rate": 1.4112903225806454e-05,
|
| 57 |
+
"loss": 2.0283,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.6412825651302605,
|
| 62 |
+
"grad_norm": 7.426329135894775,
|
| 63 |
+
"learning_rate": 1.6129032258064517e-05,
|
| 64 |
+
"loss": 1.9125,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.7214428857715431,
|
| 69 |
+
"grad_norm": 8.441859245300293,
|
| 70 |
+
"learning_rate": 1.8145161290322583e-05,
|
| 71 |
+
"loss": 1.7735,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.8016032064128257,
|
| 76 |
+
"grad_norm": 5.191440582275391,
|
| 77 |
+
"learning_rate": 2.0161290322580645e-05,
|
| 78 |
+
"loss": 1.6402,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.8817635270541082,
|
| 83 |
+
"grad_norm": 6.325778007507324,
|
| 84 |
+
"learning_rate": 2.217741935483871e-05,
|
| 85 |
+
"loss": 1.5225,
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.9619238476953907,
|
| 90 |
+
"grad_norm": 7.641424655914307,
|
| 91 |
+
"learning_rate": 2.4193548387096777e-05,
|
| 92 |
+
"loss": 1.4287,
|
| 93 |
+
"step": 120
|
|
|
|
|
|
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.9939879759519038,
|
| 97 |
+
"eval_accuracy": 0.5851183765501691,
|
| 98 |
+
"eval_loss": 1.2978211641311646,
|
| 99 |
+
"eval_runtime": 23.4495,
|
| 100 |
+
"eval_samples_per_second": 75.652,
|
| 101 |
+
"eval_steps_per_second": 2.388,
|
| 102 |
+
"step": 124
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 1.0420841683366733,
|
| 106 |
+
"grad_norm": 9.333320617675781,
|
| 107 |
+
"learning_rate": 2.620967741935484e-05,
|
| 108 |
+
"loss": 1.3617,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 1.122244488977956,
|
| 113 |
+
"grad_norm": 10.036263465881348,
|
| 114 |
+
"learning_rate": 2.822580645161291e-05,
|
| 115 |
+
"loss": 1.3084,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 1.2024048096192386,
|
| 120 |
+
"grad_norm": 11.795063018798828,
|
| 121 |
+
"learning_rate": 3.024193548387097e-05,
|
| 122 |
+
"loss": 1.2472,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 1.282565130260521,
|
| 127 |
+
"grad_norm": 11.583420753479004,
|
| 128 |
+
"learning_rate": 3.2258064516129034e-05,
|
| 129 |
+
"loss": 1.1335,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 1.3627254509018036,
|
| 134 |
+
"grad_norm": 14.882524490356445,
|
| 135 |
+
"learning_rate": 3.427419354838709e-05,
|
| 136 |
+
"loss": 1.102,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 1.4428857715430863,
|
| 141 |
+
"grad_norm": 14.157336235046387,
|
| 142 |
+
"learning_rate": 3.6290322580645165e-05,
|
| 143 |
+
"loss": 1.049,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 1.5230460921843687,
|
| 148 |
+
"grad_norm": 10.484189987182617,
|
| 149 |
+
"learning_rate": 3.8306451612903224e-05,
|
| 150 |
+
"loss": 1.0445,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
+
"epoch": 1.6032064128256514,
|
| 155 |
+
"grad_norm": 14.128747940063477,
|
| 156 |
+
"learning_rate": 4.032258064516129e-05,
|
| 157 |
+
"loss": 0.995,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 1.6833667334669338,
|
| 162 |
+
"grad_norm": 9.768001556396484,
|
| 163 |
+
"learning_rate": 4.2338709677419356e-05,
|
| 164 |
+
"loss": 0.9573,
|
| 165 |
"step": 210
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 1.7635270541082164,
|
| 169 |
+
"grad_norm": 13.823319435119629,
|
| 170 |
+
"learning_rate": 4.435483870967742e-05,
|
| 171 |
+
"loss": 0.9072,
|
| 172 |
"step": 220
|
| 173 |
},
|
| 174 |
{
|
| 175 |
+
"epoch": 1.843687374749499,
|
| 176 |
+
"grad_norm": 9.132129669189453,
|
| 177 |
+
"learning_rate": 4.637096774193548e-05,
|
| 178 |
+
"loss": 0.8156,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 179 |
"step": 230
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 1.9238476953907817,
|
| 183 |
+
"grad_norm": 18.744123458862305,
|
| 184 |
+
"learning_rate": 4.8387096774193554e-05,
|
| 185 |
+
"loss": 0.8329,
|
| 186 |
"step": 240
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 1.9959919839679359,
|
| 190 |
+
"eval_accuracy": 0.7728297632468997,
|
| 191 |
+
"eval_loss": 0.6986980438232422,
|
| 192 |
+
"eval_runtime": 11.7756,
|
| 193 |
+
"eval_samples_per_second": 150.65,
|
| 194 |
+
"eval_steps_per_second": 4.756,
|
| 195 |
+
"step": 249
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.004008016032064,
|
| 199 |
+
"grad_norm": 8.23229694366455,
|
| 200 |
+
"learning_rate": 4.995519713261649e-05,
|
| 201 |
+
"loss": 0.8108,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 2.0841683366733466,
|
| 206 |
+
"grad_norm": 13.277573585510254,
|
| 207 |
+
"learning_rate": 4.973118279569893e-05,
|
| 208 |
+
"loss": 0.7733,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 2.164328657314629,
|
| 213 |
+
"grad_norm": 13.641548156738281,
|
| 214 |
+
"learning_rate": 4.950716845878137e-05,
|
| 215 |
+
"loss": 0.7904,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 2.244488977955912,
|
| 220 |
+
"grad_norm": 18.69782829284668,
|
| 221 |
+
"learning_rate": 4.92831541218638e-05,
|
| 222 |
+
"loss": 0.7419,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 2.3246492985971945,
|
| 227 |
+
"grad_norm": 15.437073707580566,
|
| 228 |
+
"learning_rate": 4.905913978494624e-05,
|
| 229 |
+
"loss": 0.7647,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 2.404809619238477,
|
| 234 |
+
"grad_norm": 21.065357208251953,
|
| 235 |
+
"learning_rate": 4.8835125448028677e-05,
|
| 236 |
+
"loss": 0.7303,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 2.4849699398797593,
|
| 241 |
+
"grad_norm": 16.6332950592041,
|
| 242 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 243 |
+
"loss": 0.764,
|
| 244 |
"step": 310
|
| 245 |
},
|
| 246 |
{
|
| 247 |
+
"epoch": 2.565130260521042,
|
| 248 |
+
"grad_norm": 13.331892967224121,
|
| 249 |
+
"learning_rate": 4.8387096774193554e-05,
|
| 250 |
+
"loss": 0.7586,
|
| 251 |
"step": 320
|
| 252 |
},
|
| 253 |
{
|
| 254 |
+
"epoch": 2.6452905811623246,
|
| 255 |
+
"grad_norm": 18.221023559570312,
|
| 256 |
+
"learning_rate": 4.8163082437275986e-05,
|
| 257 |
+
"loss": 0.7078,
|
| 258 |
"step": 330
|
| 259 |
},
|
| 260 |
{
|
| 261 |
+
"epoch": 2.7254509018036073,
|
| 262 |
+
"grad_norm": 16.339580535888672,
|
| 263 |
+
"learning_rate": 4.7939068100358424e-05,
|
| 264 |
+
"loss": 0.7008,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 265 |
"step": 340
|
| 266 |
},
|
| 267 |
{
|
| 268 |
+
"epoch": 2.80561122244489,
|
| 269 |
+
"grad_norm": 12.270729064941406,
|
| 270 |
+
"learning_rate": 4.771505376344086e-05,
|
| 271 |
+
"loss": 0.7941,
|
| 272 |
"step": 350
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 2.8857715430861726,
|
| 276 |
+
"grad_norm": 10.448567390441895,
|
| 277 |
+
"learning_rate": 4.74910394265233e-05,
|
| 278 |
+
"loss": 0.6575,
|
| 279 |
"step": 360
|
| 280 |
},
|
| 281 |
{
|
| 282 |
+
"epoch": 2.9659318637274548,
|
| 283 |
+
"grad_norm": 12.076117515563965,
|
| 284 |
+
"learning_rate": 4.726702508960574e-05,
|
| 285 |
+
"loss": 0.7348,
|
| 286 |
"step": 370
|
| 287 |
},
|
| 288 |
{
|
| 289 |
+
"epoch": 2.997995991983968,
|
| 290 |
+
"eval_accuracy": 0.8179255918827508,
|
| 291 |
+
"eval_loss": 0.5658715963363647,
|
| 292 |
+
"eval_runtime": 12.6995,
|
| 293 |
+
"eval_samples_per_second": 139.691,
|
| 294 |
+
"eval_steps_per_second": 4.41,
|
| 295 |
+
"step": 374
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.0460921843687374,
|
| 299 |
+
"grad_norm": 10.130627632141113,
|
| 300 |
+
"learning_rate": 4.704301075268818e-05,
|
| 301 |
+
"loss": 0.6752,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 3.12625250501002,
|
| 306 |
+
"grad_norm": 15.763223648071289,
|
| 307 |
+
"learning_rate": 4.681899641577061e-05,
|
| 308 |
+
"loss": 0.6367,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 3.2064128256513027,
|
| 313 |
+
"grad_norm": 13.580248832702637,
|
| 314 |
+
"learning_rate": 4.659498207885305e-05,
|
| 315 |
+
"loss": 0.6239,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
+
"epoch": 3.2865731462925853,
|
| 320 |
+
"grad_norm": 11.084284782409668,
|
| 321 |
+
"learning_rate": 4.637096774193548e-05,
|
| 322 |
+
"loss": 0.6857,
|
| 323 |
"step": 410
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"epoch": 3.3667334669338675,
|
| 327 |
+
"grad_norm": 12.55604362487793,
|
| 328 |
+
"learning_rate": 4.614695340501792e-05,
|
| 329 |
+
"loss": 0.6603,
|
| 330 |
"step": 420
|
| 331 |
},
|
| 332 |
{
|
| 333 |
+
"epoch": 3.44689378757515,
|
| 334 |
+
"grad_norm": 9.201930046081543,
|
| 335 |
+
"learning_rate": 4.5922939068100365e-05,
|
| 336 |
+
"loss": 0.6514,
|
| 337 |
"step": 430
|
| 338 |
},
|
| 339 |
{
|
| 340 |
+
"epoch": 3.527054108216433,
|
| 341 |
+
"grad_norm": 12.60409927368164,
|
| 342 |
+
"learning_rate": 4.56989247311828e-05,
|
| 343 |
+
"loss": 0.6237,
|
| 344 |
"step": 440
|
| 345 |
},
|
| 346 |
{
|
| 347 |
+
"epoch": 3.6072144288577155,
|
| 348 |
+
"grad_norm": 21.13867950439453,
|
| 349 |
+
"learning_rate": 4.5474910394265236e-05,
|
| 350 |
+
"loss": 0.6698,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 351 |
"step": 450
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 3.687374749498998,
|
| 355 |
+
"grad_norm": 15.415425300598145,
|
| 356 |
+
"learning_rate": 4.5250896057347674e-05,
|
| 357 |
+
"loss": 0.6016,
|
| 358 |
"step": 460
|
| 359 |
},
|
| 360 |
{
|
| 361 |
+
"epoch": 3.7675350701402808,
|
| 362 |
+
"grad_norm": 10.865119934082031,
|
| 363 |
+
"learning_rate": 4.5026881720430106e-05,
|
| 364 |
+
"loss": 0.633,
|
| 365 |
"step": 470
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 3.847695390781563,
|
| 369 |
+
"grad_norm": 12.04747200012207,
|
| 370 |
+
"learning_rate": 4.4802867383512545e-05,
|
| 371 |
+
"loss": 0.686,
|
| 372 |
"step": 480
|
| 373 |
},
|
| 374 |
{
|
| 375 |
+
"epoch": 3.9278557114228456,
|
| 376 |
+
"grad_norm": 15.560898780822754,
|
| 377 |
+
"learning_rate": 4.4578853046594983e-05,
|
| 378 |
+
"loss": 0.611,
|
| 379 |
"step": 490
|
| 380 |
},
|
| 381 |
{
|
| 382 |
+
"epoch": 4.0,
|
| 383 |
+
"eval_accuracy": 0.8297632468996617,
|
| 384 |
+
"eval_loss": 0.5378695726394653,
|
| 385 |
+
"eval_runtime": 12.4744,
|
| 386 |
+
"eval_samples_per_second": 142.212,
|
| 387 |
+
"eval_steps_per_second": 4.489,
|
| 388 |
+
"step": 499
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.008016032064128,
|
| 392 |
+
"grad_norm": 12.36653995513916,
|
| 393 |
+
"learning_rate": 4.435483870967742e-05,
|
| 394 |
+
"loss": 0.6783,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 4.0881763527054105,
|
| 399 |
+
"grad_norm": 12.612045288085938,
|
| 400 |
+
"learning_rate": 4.413082437275986e-05,
|
| 401 |
+
"loss": 0.6591,
|
| 402 |
"step": 510
|
| 403 |
},
|
| 404 |
{
|
| 405 |
+
"epoch": 4.168336673346693,
|
| 406 |
+
"grad_norm": 9.516498565673828,
|
| 407 |
+
"learning_rate": 4.390681003584229e-05,
|
| 408 |
+
"loss": 0.6409,
|
| 409 |
"step": 520
|
| 410 |
},
|
| 411 |
{
|
| 412 |
+
"epoch": 4.248496993987976,
|
| 413 |
+
"grad_norm": 10.227922439575195,
|
| 414 |
+
"learning_rate": 4.368279569892473e-05,
|
| 415 |
+
"loss": 0.6113,
|
| 416 |
"step": 530
|
| 417 |
},
|
| 418 |
{
|
| 419 |
+
"epoch": 4.328657314629258,
|
| 420 |
+
"grad_norm": 9.941215515136719,
|
| 421 |
+
"learning_rate": 4.345878136200717e-05,
|
| 422 |
+
"loss": 0.5545,
|
| 423 |
"step": 540
|
| 424 |
},
|
| 425 |
{
|
| 426 |
+
"epoch": 4.408817635270541,
|
| 427 |
+
"grad_norm": 11.132833480834961,
|
| 428 |
+
"learning_rate": 4.323476702508961e-05,
|
| 429 |
+
"loss": 0.5876,
|
| 430 |
"step": 550
|
| 431 |
},
|
| 432 |
{
|
| 433 |
+
"epoch": 4.488977955911824,
|
| 434 |
+
"grad_norm": 17.30998992919922,
|
| 435 |
+
"learning_rate": 4.301075268817205e-05,
|
| 436 |
+
"loss": 0.6059,
|
| 437 |
"step": 560
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 4.569138276553106,
|
| 441 |
+
"grad_norm": 12.370113372802734,
|
| 442 |
+
"learning_rate": 4.2786738351254486e-05,
|
| 443 |
+
"loss": 0.572,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 444 |
"step": 570
|
| 445 |
},
|
| 446 |
{
|
| 447 |
+
"epoch": 4.649298597194389,
|
| 448 |
+
"grad_norm": 8.4649019241333,
|
| 449 |
+
"learning_rate": 4.256272401433692e-05,
|
| 450 |
+
"loss": 0.5474,
|
| 451 |
"step": 580
|
| 452 |
},
|
| 453 |
{
|
| 454 |
+
"epoch": 4.729458917835672,
|
| 455 |
+
"grad_norm": 13.911017417907715,
|
| 456 |
+
"learning_rate": 4.2338709677419356e-05,
|
| 457 |
+
"loss": 0.5745,
|
| 458 |
"step": 590
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 4.809619238476954,
|
| 462 |
+
"grad_norm": 13.061511993408203,
|
| 463 |
+
"learning_rate": 4.2114695340501795e-05,
|
| 464 |
+
"loss": 0.6151,
|
| 465 |
"step": 600
|
| 466 |
},
|
| 467 |
{
|
| 468 |
+
"epoch": 4.889779559118237,
|
| 469 |
+
"grad_norm": 17.543981552124023,
|
| 470 |
+
"learning_rate": 4.1890681003584233e-05,
|
| 471 |
+
"loss": 0.5702,
|
| 472 |
"step": 610
|
| 473 |
},
|
| 474 |
{
|
| 475 |
+
"epoch": 4.969939879759519,
|
| 476 |
+
"grad_norm": 14.352049827575684,
|
| 477 |
+
"learning_rate": 4.166666666666667e-05,
|
| 478 |
+
"loss": 0.5929,
|
| 479 |
"step": 620
|
| 480 |
},
|
| 481 |
{
|
| 482 |
+
"epoch": 4.993987975951904,
|
| 483 |
+
"eval_accuracy": 0.8376550169109357,
|
| 484 |
+
"eval_loss": 0.4972485899925232,
|
| 485 |
+
"eval_runtime": 30.5452,
|
| 486 |
+
"eval_samples_per_second": 58.078,
|
| 487 |
+
"eval_steps_per_second": 1.833,
|
| 488 |
+
"step": 623
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 5.050100200400801,
|
| 492 |
+
"grad_norm": 12.319650650024414,
|
| 493 |
+
"learning_rate": 4.1442652329749104e-05,
|
| 494 |
+
"loss": 0.6066,
|
| 495 |
"step": 630
|
| 496 |
},
|
| 497 |
{
|
| 498 |
+
"epoch": 5.130260521042084,
|
| 499 |
+
"grad_norm": 13.590569496154785,
|
| 500 |
+
"learning_rate": 4.121863799283154e-05,
|
| 501 |
+
"loss": 0.561,
|
| 502 |
"step": 640
|
| 503 |
},
|
| 504 |
{
|
| 505 |
+
"epoch": 5.210420841683367,
|
| 506 |
+
"grad_norm": 6.548098087310791,
|
| 507 |
+
"learning_rate": 4.099462365591398e-05,
|
| 508 |
+
"loss": 0.5747,
|
| 509 |
"step": 650
|
| 510 |
},
|
| 511 |
{
|
| 512 |
+
"epoch": 5.290581162324649,
|
| 513 |
+
"grad_norm": 11.081100463867188,
|
| 514 |
+
"learning_rate": 4.077060931899642e-05,
|
| 515 |
+
"loss": 0.5469,
|
| 516 |
"step": 660
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"epoch": 5.370741482965932,
|
| 520 |
+
"grad_norm": 11.028233528137207,
|
| 521 |
+
"learning_rate": 4.054659498207886e-05,
|
| 522 |
+
"loss": 0.5913,
|
| 523 |
"step": 670
|
| 524 |
},
|
| 525 |
{
|
| 526 |
+
"epoch": 5.4509018036072145,
|
| 527 |
+
"grad_norm": 16.77172088623047,
|
| 528 |
+
"learning_rate": 4.032258064516129e-05,
|
| 529 |
+
"loss": 0.5454,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 530 |
"step": 680
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 5.531062124248497,
|
| 534 |
+
"grad_norm": 12.664057731628418,
|
| 535 |
+
"learning_rate": 4.009856630824373e-05,
|
| 536 |
+
"loss": 0.5376,
|
| 537 |
"step": 690
|
| 538 |
},
|
| 539 |
{
|
| 540 |
+
"epoch": 5.61122244488978,
|
| 541 |
+
"grad_norm": 16.970354080200195,
|
| 542 |
+
"learning_rate": 3.987455197132617e-05,
|
| 543 |
+
"loss": 0.6195,
|
| 544 |
"step": 700
|
| 545 |
},
|
| 546 |
{
|
| 547 |
+
"epoch": 5.6913827655310625,
|
| 548 |
+
"grad_norm": 20.813093185424805,
|
| 549 |
+
"learning_rate": 3.96505376344086e-05,
|
| 550 |
+
"loss": 0.5759,
|
| 551 |
"step": 710
|
| 552 |
},
|
| 553 |
{
|
| 554 |
+
"epoch": 5.771543086172345,
|
| 555 |
+
"grad_norm": 12.778873443603516,
|
| 556 |
+
"learning_rate": 3.9426523297491045e-05,
|
| 557 |
+
"loss": 0.5636,
|
| 558 |
"step": 720
|
| 559 |
},
|
| 560 |
{
|
| 561 |
+
"epoch": 5.851703406813627,
|
| 562 |
+
"grad_norm": 9.49085521697998,
|
| 563 |
+
"learning_rate": 3.9202508960573483e-05,
|
| 564 |
+
"loss": 0.5213,
|
| 565 |
"step": 730
|
| 566 |
},
|
| 567 |
{
|
| 568 |
+
"epoch": 5.9318637274549095,
|
| 569 |
+
"grad_norm": 16.53606605529785,
|
| 570 |
+
"learning_rate": 3.8978494623655915e-05,
|
| 571 |
+
"loss": 0.5227,
|
| 572 |
"step": 740
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 5.995991983967936,
|
| 576 |
+
"eval_accuracy": 0.8478015783540023,
|
| 577 |
+
"eval_loss": 0.4714604616165161,
|
| 578 |
+
"eval_runtime": 11.6058,
|
| 579 |
+
"eval_samples_per_second": 152.855,
|
| 580 |
+
"eval_steps_per_second": 4.825,
|
| 581 |
+
"step": 748
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 6.012024048096192,
|
| 585 |
+
"grad_norm": 15.601022720336914,
|
| 586 |
+
"learning_rate": 3.8754480286738354e-05,
|
| 587 |
+
"loss": 0.5287,
|
| 588 |
"step": 750
|
| 589 |
},
|
| 590 |
{
|
| 591 |
+
"epoch": 6.092184368737475,
|
| 592 |
+
"grad_norm": 10.821127891540527,
|
| 593 |
+
"learning_rate": 3.8530465949820786e-05,
|
| 594 |
+
"loss": 0.5278,
|
| 595 |
"step": 760
|
| 596 |
},
|
| 597 |
{
|
| 598 |
+
"epoch": 6.1723446893787575,
|
| 599 |
+
"grad_norm": 15.051136016845703,
|
| 600 |
+
"learning_rate": 3.8306451612903224e-05,
|
| 601 |
+
"loss": 0.5491,
|
| 602 |
"step": 770
|
| 603 |
},
|
| 604 |
{
|
| 605 |
+
"epoch": 6.25250501002004,
|
| 606 |
+
"grad_norm": 13.785345077514648,
|
| 607 |
+
"learning_rate": 3.808243727598566e-05,
|
| 608 |
+
"loss": 0.4941,
|
| 609 |
"step": 780
|
| 610 |
},
|
| 611 |
{
|
| 612 |
+
"epoch": 6.332665330661323,
|
| 613 |
+
"grad_norm": 12.571328163146973,
|
| 614 |
+
"learning_rate": 3.78584229390681e-05,
|
| 615 |
+
"loss": 0.4668,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 616 |
"step": 790
|
| 617 |
},
|
| 618 |
{
|
| 619 |
+
"epoch": 6.412825651302605,
|
| 620 |
+
"grad_norm": 14.443199157714844,
|
| 621 |
+
"learning_rate": 3.763440860215054e-05,
|
| 622 |
+
"loss": 0.5502,
|
| 623 |
"step": 800
|
| 624 |
},
|
| 625 |
{
|
| 626 |
+
"epoch": 6.492985971943888,
|
| 627 |
+
"grad_norm": 12.781950950622559,
|
| 628 |
+
"learning_rate": 3.741039426523298e-05,
|
| 629 |
+
"loss": 0.5049,
|
| 630 |
"step": 810
|
| 631 |
},
|
| 632 |
{
|
| 633 |
+
"epoch": 6.573146292585171,
|
| 634 |
+
"grad_norm": 8.832810401916504,
|
| 635 |
+
"learning_rate": 3.718637992831541e-05,
|
| 636 |
+
"loss": 0.5654,
|
| 637 |
"step": 820
|
| 638 |
},
|
| 639 |
{
|
| 640 |
+
"epoch": 6.653306613226453,
|
| 641 |
+
"grad_norm": 13.026018142700195,
|
| 642 |
+
"learning_rate": 3.696236559139785e-05,
|
| 643 |
+
"loss": 0.5289,
|
| 644 |
"step": 830
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 6.733466933867735,
|
| 648 |
+
"grad_norm": 11.173068046569824,
|
| 649 |
+
"learning_rate": 3.673835125448029e-05,
|
| 650 |
+
"loss": 0.5262,
|
| 651 |
"step": 840
|
| 652 |
},
|
| 653 |
{
|
| 654 |
+
"epoch": 6.813627254509018,
|
| 655 |
+
"grad_norm": 15.73713207244873,
|
| 656 |
+
"learning_rate": 3.651433691756273e-05,
|
| 657 |
+
"loss": 0.5239,
|
| 658 |
"step": 850
|
| 659 |
},
|
| 660 |
{
|
| 661 |
+
"epoch": 6.8937875751503,
|
| 662 |
+
"grad_norm": 11.182281494140625,
|
| 663 |
+
"learning_rate": 3.6290322580645165e-05,
|
| 664 |
+
"loss": 0.5269,
|
| 665 |
"step": 860
|
| 666 |
},
|
| 667 |
{
|
| 668 |
+
"epoch": 6.973947895791583,
|
| 669 |
+
"grad_norm": 11.750397682189941,
|
| 670 |
+
"learning_rate": 3.60663082437276e-05,
|
| 671 |
+
"loss": 0.5166,
|
| 672 |
"step": 870
|
| 673 |
},
|
| 674 |
{
|
| 675 |
+
"epoch": 6.997995991983968,
|
| 676 |
+
"eval_accuracy": 0.8494926719278467,
|
| 677 |
+
"eval_loss": 0.47609812021255493,
|
| 678 |
+
"eval_runtime": 22.6677,
|
| 679 |
+
"eval_samples_per_second": 78.261,
|
| 680 |
+
"eval_steps_per_second": 2.47,
|
| 681 |
+
"step": 873
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 7.054108216432866,
|
| 685 |
+
"grad_norm": 13.769631385803223,
|
| 686 |
+
"learning_rate": 3.5842293906810036e-05,
|
| 687 |
+
"loss": 0.5207,
|
| 688 |
"step": 880
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 7.134268537074148,
|
| 692 |
+
"grad_norm": 13.10180377960205,
|
| 693 |
+
"learning_rate": 3.5618279569892474e-05,
|
| 694 |
+
"loss": 0.5231,
|
| 695 |
"step": 890
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 7.214428857715431,
|
| 699 |
+
"grad_norm": 9.971457481384277,
|
| 700 |
+
"learning_rate": 3.539426523297491e-05,
|
| 701 |
+
"loss": 0.518,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 702 |
"step": 900
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 7.294589178356714,
|
| 706 |
+
"grad_norm": 12.092657089233398,
|
| 707 |
+
"learning_rate": 3.517025089605735e-05,
|
| 708 |
+
"loss": 0.5034,
|
| 709 |
"step": 910
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 7.374749498997996,
|
| 713 |
+
"grad_norm": 19.348663330078125,
|
| 714 |
+
"learning_rate": 3.494623655913979e-05,
|
| 715 |
+
"loss": 0.4927,
|
| 716 |
"step": 920
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 7.454909819639279,
|
| 720 |
+
"grad_norm": 10.206799507141113,
|
| 721 |
+
"learning_rate": 3.472222222222222e-05,
|
| 722 |
+
"loss": 0.5062,
|
| 723 |
"step": 930
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 7.5350701402805615,
|
| 727 |
+
"grad_norm": 9.899465560913086,
|
| 728 |
+
"learning_rate": 3.449820788530466e-05,
|
| 729 |
+
"loss": 0.5037,
|
| 730 |
"step": 940
|
| 731 |
},
|
| 732 |
{
|
| 733 |
+
"epoch": 7.615230460921843,
|
| 734 |
+
"grad_norm": 13.186443328857422,
|
| 735 |
+
"learning_rate": 3.427419354838709e-05,
|
| 736 |
+
"loss": 0.5159,
|
| 737 |
"step": 950
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 7.695390781563126,
|
| 741 |
+
"grad_norm": 9.82767391204834,
|
| 742 |
+
"learning_rate": 3.405017921146954e-05,
|
| 743 |
+
"loss": 0.479,
|
| 744 |
"step": 960
|
| 745 |
},
|
| 746 |
{
|
| 747 |
+
"epoch": 7.775551102204409,
|
| 748 |
+
"grad_norm": 9.09422492980957,
|
| 749 |
+
"learning_rate": 3.382616487455198e-05,
|
| 750 |
+
"loss": 0.5146,
|
| 751 |
"step": 970
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"epoch": 7.855711422845691,
|
| 755 |
+
"grad_norm": 11.1051025390625,
|
| 756 |
+
"learning_rate": 3.360215053763441e-05,
|
| 757 |
+
"loss": 0.4908,
|
| 758 |
"step": 980
|
| 759 |
},
|
| 760 |
{
|
| 761 |
+
"epoch": 7.935871743486974,
|
| 762 |
+
"grad_norm": 9.16980266571045,
|
| 763 |
+
"learning_rate": 3.337813620071685e-05,
|
| 764 |
+
"loss": 0.4992,
|
| 765 |
"step": 990
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 8.0,
|
| 769 |
+
"eval_accuracy": 0.8562570462232244,
|
| 770 |
+
"eval_loss": 0.432047575712204,
|
| 771 |
+
"eval_runtime": 17.8346,
|
| 772 |
+
"eval_samples_per_second": 99.469,
|
| 773 |
+
"eval_steps_per_second": 3.14,
|
| 774 |
+
"step": 998
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 8.016032064128256,
|
| 778 |
+
"grad_norm": 8.514852523803711,
|
| 779 |
+
"learning_rate": 3.3154121863799286e-05,
|
| 780 |
+
"loss": 0.4702,
|
| 781 |
"step": 1000
|
| 782 |
},
|
| 783 |
{
|
| 784 |
+
"epoch": 8.09619238476954,
|
| 785 |
+
"grad_norm": 10.90427017211914,
|
| 786 |
+
"learning_rate": 3.293010752688172e-05,
|
| 787 |
+
"loss": 0.4809,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 788 |
"step": 1010
|
| 789 |
},
|
| 790 |
{
|
| 791 |
+
"epoch": 8.176352705410821,
|
| 792 |
+
"grad_norm": 13.75596809387207,
|
| 793 |
+
"learning_rate": 3.270609318996416e-05,
|
| 794 |
+
"loss": 0.4769,
|
| 795 |
"step": 1020
|
| 796 |
},
|
| 797 |
{
|
| 798 |
+
"epoch": 8.256513026052104,
|
| 799 |
+
"grad_norm": 14.506204605102539,
|
| 800 |
+
"learning_rate": 3.24820788530466e-05,
|
| 801 |
+
"loss": 0.5157,
|
| 802 |
"step": 1030
|
| 803 |
},
|
| 804 |
{
|
| 805 |
+
"epoch": 8.336673346693386,
|
| 806 |
+
"grad_norm": 10.074383735656738,
|
| 807 |
+
"learning_rate": 3.2258064516129034e-05,
|
| 808 |
+
"loss": 0.4916,
|
| 809 |
"step": 1040
|
| 810 |
},
|
| 811 |
{
|
| 812 |
+
"epoch": 8.41683366733467,
|
| 813 |
+
"grad_norm": 13.06877326965332,
|
| 814 |
+
"learning_rate": 3.203405017921147e-05,
|
| 815 |
+
"loss": 0.4953,
|
| 816 |
"step": 1050
|
| 817 |
},
|
| 818 |
{
|
| 819 |
+
"epoch": 8.496993987975952,
|
| 820 |
+
"grad_norm": 7.9596171379089355,
|
| 821 |
+
"learning_rate": 3.1810035842293904e-05,
|
| 822 |
+
"loss": 0.4879,
|
| 823 |
"step": 1060
|
| 824 |
},
|
| 825 |
{
|
| 826 |
+
"epoch": 8.577154308617235,
|
| 827 |
+
"grad_norm": 11.05156135559082,
|
| 828 |
+
"learning_rate": 3.158602150537634e-05,
|
| 829 |
+
"loss": 0.4397,
|
| 830 |
"step": 1070
|
| 831 |
},
|
| 832 |
{
|
| 833 |
+
"epoch": 8.657314629258517,
|
| 834 |
+
"grad_norm": 9.935453414916992,
|
| 835 |
+
"learning_rate": 3.136200716845878e-05,
|
| 836 |
+
"loss": 0.5141,
|
| 837 |
"step": 1080
|
| 838 |
},
|
| 839 |
{
|
| 840 |
+
"epoch": 8.7374749498998,
|
| 841 |
+
"grad_norm": 9.928804397583008,
|
| 842 |
+
"learning_rate": 3.113799283154122e-05,
|
| 843 |
+
"loss": 0.4781,
|
| 844 |
"step": 1090
|
| 845 |
},
|
| 846 |
{
|
| 847 |
+
"epoch": 8.817635270541082,
|
| 848 |
+
"grad_norm": 7.301691055297852,
|
| 849 |
+
"learning_rate": 3.091397849462366e-05,
|
| 850 |
+
"loss": 0.4484,
|
| 851 |
"step": 1100
|
| 852 |
},
|
| 853 |
{
|
| 854 |
+
"epoch": 8.897795591182366,
|
| 855 |
+
"grad_norm": 13.609901428222656,
|
| 856 |
+
"learning_rate": 3.06899641577061e-05,
|
| 857 |
+
"loss": 0.4656,
|
| 858 |
"step": 1110
|
| 859 |
},
|
| 860 |
{
|
| 861 |
+
"epoch": 8.977955911823647,
|
| 862 |
+
"grad_norm": 10.269015312194824,
|
| 863 |
+
"learning_rate": 3.046594982078853e-05,
|
| 864 |
+
"loss": 0.4528,
|
| 865 |
"step": 1120
|
| 866 |
},
|
| 867 |
{
|
| 868 |
+
"epoch": 8.993987975951903,
|
| 869 |
+
"eval_accuracy": 0.8641488162344984,
|
| 870 |
+
"eval_loss": 0.4410019814968109,
|
| 871 |
+
"eval_runtime": 27.5679,
|
| 872 |
+
"eval_samples_per_second": 64.35,
|
| 873 |
+
"eval_steps_per_second": 2.031,
|
| 874 |
+
"step": 1122
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 9.05811623246493,
|
| 878 |
+
"grad_norm": 8.7766695022583,
|
| 879 |
+
"learning_rate": 3.024193548387097e-05,
|
| 880 |
+
"loss": 0.4843,
|
| 881 |
+
"step": 1130
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 9.138276553106213,
|
| 885 |
+
"grad_norm": 10.279314994812012,
|
| 886 |
+
"learning_rate": 3.0017921146953403e-05,
|
| 887 |
+
"loss": 0.5017,
|
| 888 |
+
"step": 1140
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 9.218436873747494,
|
| 892 |
+
"grad_norm": 13.840995788574219,
|
| 893 |
+
"learning_rate": 2.979390681003584e-05,
|
| 894 |
+
"loss": 0.4443,
|
| 895 |
+
"step": 1150
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 9.298597194388778,
|
| 899 |
+
"grad_norm": 14.21786117553711,
|
| 900 |
+
"learning_rate": 2.9569892473118284e-05,
|
| 901 |
+
"loss": 0.4319,
|
| 902 |
+
"step": 1160
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 9.37875751503006,
|
| 906 |
+
"grad_norm": 9.682762145996094,
|
| 907 |
+
"learning_rate": 2.9345878136200715e-05,
|
| 908 |
+
"loss": 0.4692,
|
| 909 |
+
"step": 1170
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 9.458917835671343,
|
| 913 |
+
"grad_norm": 12.985733985900879,
|
| 914 |
+
"learning_rate": 2.9121863799283154e-05,
|
| 915 |
+
"loss": 0.4221,
|
| 916 |
+
"step": 1180
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 9.539078156312625,
|
| 920 |
+
"grad_norm": 12.35405445098877,
|
| 921 |
+
"learning_rate": 2.8897849462365596e-05,
|
| 922 |
+
"loss": 0.4835,
|
| 923 |
+
"step": 1190
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 9.619238476953909,
|
| 927 |
+
"grad_norm": 7.067807197570801,
|
| 928 |
+
"learning_rate": 2.8673835125448028e-05,
|
| 929 |
+
"loss": 0.4548,
|
| 930 |
+
"step": 1200
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 9.69939879759519,
|
| 934 |
+
"grad_norm": 10.279123306274414,
|
| 935 |
+
"learning_rate": 2.8449820788530467e-05,
|
| 936 |
+
"loss": 0.4791,
|
| 937 |
+
"step": 1210
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 9.779559118236474,
|
| 941 |
+
"grad_norm": 12.814294815063477,
|
| 942 |
+
"learning_rate": 2.822580645161291e-05,
|
| 943 |
+
"loss": 0.4551,
|
| 944 |
+
"step": 1220
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 9.859719438877756,
|
| 948 |
+
"grad_norm": 15.132489204406738,
|
| 949 |
+
"learning_rate": 2.800179211469534e-05,
|
| 950 |
+
"loss": 0.3957,
|
| 951 |
+
"step": 1230
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 9.939879759519037,
|
| 955 |
+
"grad_norm": 11.959942817687988,
|
| 956 |
+
"learning_rate": 2.777777777777778e-05,
|
| 957 |
+
"loss": 0.4566,
|
| 958 |
+
"step": 1240
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 9.995991983967937,
|
| 962 |
+
"eval_accuracy": 0.8641488162344984,
|
| 963 |
+
"eval_loss": 0.42970511317253113,
|
| 964 |
+
"eval_runtime": 16.9001,
|
| 965 |
+
"eval_samples_per_second": 104.97,
|
| 966 |
+
"eval_steps_per_second": 3.314,
|
| 967 |
+
"step": 1247
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 10.02004008016032,
|
| 971 |
+
"grad_norm": 12.989178657531738,
|
| 972 |
+
"learning_rate": 2.7553763440860214e-05,
|
| 973 |
+
"loss": 0.4402,
|
| 974 |
+
"step": 1250
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 10.100200400801603,
|
| 978 |
+
"grad_norm": 11.985700607299805,
|
| 979 |
+
"learning_rate": 2.7329749103942653e-05,
|
| 980 |
+
"loss": 0.4544,
|
| 981 |
+
"step": 1260
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 10.180360721442886,
|
| 985 |
+
"grad_norm": 17.515514373779297,
|
| 986 |
+
"learning_rate": 2.710573476702509e-05,
|
| 987 |
+
"loss": 0.3676,
|
| 988 |
+
"step": 1270
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 10.260521042084168,
|
| 992 |
+
"grad_norm": 11.017400741577148,
|
| 993 |
+
"learning_rate": 2.6881720430107527e-05,
|
| 994 |
+
"loss": 0.4534,
|
| 995 |
+
"step": 1280
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 10.340681362725451,
|
| 999 |
+
"grad_norm": 7.939273357391357,
|
| 1000 |
+
"learning_rate": 2.6657706093189965e-05,
|
| 1001 |
+
"loss": 0.4719,
|
| 1002 |
+
"step": 1290
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 10.420841683366733,
|
| 1006 |
+
"grad_norm": 13.53430461883545,
|
| 1007 |
+
"learning_rate": 2.6433691756272404e-05,
|
| 1008 |
+
"loss": 0.4162,
|
| 1009 |
+
"step": 1300
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 10.501002004008017,
|
| 1013 |
+
"grad_norm": 9.599760055541992,
|
| 1014 |
+
"learning_rate": 2.620967741935484e-05,
|
| 1015 |
+
"loss": 0.46,
|
| 1016 |
+
"step": 1310
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 10.581162324649299,
|
| 1020 |
+
"grad_norm": 7.481749057769775,
|
| 1021 |
+
"learning_rate": 2.5985663082437278e-05,
|
| 1022 |
+
"loss": 0.405,
|
| 1023 |
+
"step": 1320
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 10.661322645290582,
|
| 1027 |
+
"grad_norm": 17.151025772094727,
|
| 1028 |
+
"learning_rate": 2.5761648745519713e-05,
|
| 1029 |
+
"loss": 0.4484,
|
| 1030 |
+
"step": 1330
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 10.741482965931864,
|
| 1034 |
+
"grad_norm": 11.18791389465332,
|
| 1035 |
+
"learning_rate": 2.5537634408602152e-05,
|
| 1036 |
+
"loss": 0.4461,
|
| 1037 |
+
"step": 1340
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 10.821643286573146,
|
| 1041 |
+
"grad_norm": 9.898661613464355,
|
| 1042 |
+
"learning_rate": 2.531362007168459e-05,
|
| 1043 |
+
"loss": 0.3958,
|
| 1044 |
+
"step": 1350
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 10.901803607214429,
|
| 1048 |
+
"grad_norm": 9.442924499511719,
|
| 1049 |
+
"learning_rate": 2.5089605734767026e-05,
|
| 1050 |
+
"loss": 0.4279,
|
| 1051 |
+
"step": 1360
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 10.98196392785571,
|
| 1055 |
+
"grad_norm": 11.045487403869629,
|
| 1056 |
+
"learning_rate": 2.4865591397849464e-05,
|
| 1057 |
+
"loss": 0.4294,
|
| 1058 |
+
"step": 1370
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 10.997995991983968,
|
| 1062 |
+
"eval_accuracy": 0.8607666290868095,
|
| 1063 |
+
"eval_loss": 0.42823219299316406,
|
| 1064 |
+
"eval_runtime": 26.2766,
|
| 1065 |
+
"eval_samples_per_second": 67.512,
|
| 1066 |
+
"eval_steps_per_second": 2.131,
|
| 1067 |
+
"step": 1372
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 11.062124248496994,
|
| 1071 |
+
"grad_norm": 9.318482398986816,
|
| 1072 |
+
"learning_rate": 2.46415770609319e-05,
|
| 1073 |
+
"loss": 0.4314,
|
| 1074 |
+
"step": 1380
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 11.142284569138276,
|
| 1078 |
+
"grad_norm": 16.068525314331055,
|
| 1079 |
+
"learning_rate": 2.4417562724014338e-05,
|
| 1080 |
+
"loss": 0.3944,
|
| 1081 |
+
"step": 1390
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 11.22244488977956,
|
| 1085 |
+
"grad_norm": 6.959997177124023,
|
| 1086 |
+
"learning_rate": 2.4193548387096777e-05,
|
| 1087 |
+
"loss": 0.4251,
|
| 1088 |
+
"step": 1400
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 11.302605210420841,
|
| 1092 |
+
"grad_norm": 11.282358169555664,
|
| 1093 |
+
"learning_rate": 2.3969534050179212e-05,
|
| 1094 |
+
"loss": 0.4074,
|
| 1095 |
+
"step": 1410
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 11.382765531062125,
|
| 1099 |
+
"grad_norm": 8.684910774230957,
|
| 1100 |
+
"learning_rate": 2.374551971326165e-05,
|
| 1101 |
+
"loss": 0.4426,
|
| 1102 |
+
"step": 1420
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 11.462925851703407,
|
| 1106 |
+
"grad_norm": 11.480581283569336,
|
| 1107 |
+
"learning_rate": 2.352150537634409e-05,
|
| 1108 |
+
"loss": 0.413,
|
| 1109 |
+
"step": 1430
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 11.54308617234469,
|
| 1113 |
+
"grad_norm": 10.927531242370605,
|
| 1114 |
+
"learning_rate": 2.3297491039426525e-05,
|
| 1115 |
+
"loss": 0.4338,
|
| 1116 |
+
"step": 1440
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 11.623246492985972,
|
| 1120 |
+
"grad_norm": 10.118310928344727,
|
| 1121 |
+
"learning_rate": 2.307347670250896e-05,
|
| 1122 |
+
"loss": 0.4536,
|
| 1123 |
+
"step": 1450
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 11.703406813627254,
|
| 1127 |
+
"grad_norm": 10.131954193115234,
|
| 1128 |
+
"learning_rate": 2.28494623655914e-05,
|
| 1129 |
+
"loss": 0.4164,
|
| 1130 |
+
"step": 1460
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 11.783567134268537,
|
| 1134 |
+
"grad_norm": 14.58598804473877,
|
| 1135 |
+
"learning_rate": 2.2625448028673837e-05,
|
| 1136 |
+
"loss": 0.4342,
|
| 1137 |
+
"step": 1470
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 11.863727454909819,
|
| 1141 |
+
"grad_norm": 12.672148704528809,
|
| 1142 |
+
"learning_rate": 2.2401433691756272e-05,
|
| 1143 |
+
"loss": 0.4393,
|
| 1144 |
+
"step": 1480
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 11.943887775551103,
|
| 1148 |
+
"grad_norm": 11.871652603149414,
|
| 1149 |
+
"learning_rate": 2.217741935483871e-05,
|
| 1150 |
+
"loss": 0.3771,
|
| 1151 |
+
"step": 1490
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 12.0,
|
| 1155 |
+
"eval_accuracy": 0.85456595264938,
|
| 1156 |
+
"eval_loss": 0.4546312391757965,
|
| 1157 |
+
"eval_runtime": 21.2281,
|
| 1158 |
+
"eval_samples_per_second": 83.569,
|
| 1159 |
+
"eval_steps_per_second": 2.638,
|
| 1160 |
+
"step": 1497
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 12.024048096192384,
|
| 1164 |
+
"grad_norm": 12.87192440032959,
|
| 1165 |
+
"learning_rate": 2.1953405017921146e-05,
|
| 1166 |
+
"loss": 0.404,
|
| 1167 |
+
"step": 1500
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 12.104208416833668,
|
| 1171 |
+
"grad_norm": 11.67623519897461,
|
| 1172 |
+
"learning_rate": 2.1729390681003585e-05,
|
| 1173 |
+
"loss": 0.3968,
|
| 1174 |
+
"step": 1510
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 12.18436873747495,
|
| 1178 |
+
"grad_norm": 11.608409881591797,
|
| 1179 |
+
"learning_rate": 2.1505376344086024e-05,
|
| 1180 |
+
"loss": 0.3809,
|
| 1181 |
+
"step": 1520
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 12.264529058116233,
|
| 1185 |
+
"grad_norm": 9.568375587463379,
|
| 1186 |
+
"learning_rate": 2.128136200716846e-05,
|
| 1187 |
+
"loss": 0.4135,
|
| 1188 |
+
"step": 1530
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 12.344689378757515,
|
| 1192 |
+
"grad_norm": 10.64120864868164,
|
| 1193 |
+
"learning_rate": 2.1057347670250897e-05,
|
| 1194 |
+
"loss": 0.411,
|
| 1195 |
+
"step": 1540
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 12.424849699398798,
|
| 1199 |
+
"grad_norm": 9.730778694152832,
|
| 1200 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 1201 |
+
"loss": 0.3948,
|
| 1202 |
+
"step": 1550
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 12.50501002004008,
|
| 1206 |
+
"grad_norm": 11.325265884399414,
|
| 1207 |
+
"learning_rate": 2.060931899641577e-05,
|
| 1208 |
+
"loss": 0.403,
|
| 1209 |
+
"step": 1560
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 12.585170340681362,
|
| 1213 |
+
"grad_norm": 13.892471313476562,
|
| 1214 |
+
"learning_rate": 2.038530465949821e-05,
|
| 1215 |
+
"loss": 0.4393,
|
| 1216 |
+
"step": 1570
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 12.665330661322646,
|
| 1220 |
+
"grad_norm": 14.78463363647461,
|
| 1221 |
+
"learning_rate": 2.0161290322580645e-05,
|
| 1222 |
+
"loss": 0.3784,
|
| 1223 |
+
"step": 1580
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 12.745490981963927,
|
| 1227 |
+
"grad_norm": 14.130833625793457,
|
| 1228 |
+
"learning_rate": 1.9937275985663084e-05,
|
| 1229 |
+
"loss": 0.425,
|
| 1230 |
+
"step": 1590
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 12.82565130260521,
|
| 1234 |
+
"grad_norm": 8.856616973876953,
|
| 1235 |
+
"learning_rate": 1.9713261648745522e-05,
|
| 1236 |
+
"loss": 0.3804,
|
| 1237 |
+
"step": 1600
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 12.905811623246493,
|
| 1241 |
+
"grad_norm": 10.982331275939941,
|
| 1242 |
+
"learning_rate": 1.9489247311827958e-05,
|
| 1243 |
+
"loss": 0.4252,
|
| 1244 |
+
"step": 1610
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 12.985971943887776,
|
| 1248 |
+
"grad_norm": 9.618106842041016,
|
| 1249 |
+
"learning_rate": 1.9265232974910393e-05,
|
| 1250 |
+
"loss": 0.4224,
|
| 1251 |
+
"step": 1620
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 12.993987975951903,
|
| 1255 |
+
"eval_accuracy": 0.8624577226606539,
|
| 1256 |
+
"eval_loss": 0.448898583650589,
|
| 1257 |
+
"eval_runtime": 17.755,
|
| 1258 |
+
"eval_samples_per_second": 99.916,
|
| 1259 |
+
"eval_steps_per_second": 3.154,
|
| 1260 |
+
"step": 1621
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 13.066132264529058,
|
| 1264 |
+
"grad_norm": 18.403636932373047,
|
| 1265 |
+
"learning_rate": 1.904121863799283e-05,
|
| 1266 |
+
"loss": 0.4142,
|
| 1267 |
+
"step": 1630
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 13.146292585170341,
|
| 1271 |
+
"grad_norm": 12.012832641601562,
|
| 1272 |
+
"learning_rate": 1.881720430107527e-05,
|
| 1273 |
+
"loss": 0.3976,
|
| 1274 |
+
"step": 1640
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 13.226452905811623,
|
| 1278 |
+
"grad_norm": 10.503453254699707,
|
| 1279 |
+
"learning_rate": 1.8593189964157705e-05,
|
| 1280 |
+
"loss": 0.3707,
|
| 1281 |
+
"step": 1650
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 13.306613226452907,
|
| 1285 |
+
"grad_norm": 12.286235809326172,
|
| 1286 |
+
"learning_rate": 1.8369175627240144e-05,
|
| 1287 |
+
"loss": 0.4056,
|
| 1288 |
+
"step": 1660
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 13.386773547094188,
|
| 1292 |
+
"grad_norm": 9.312376976013184,
|
| 1293 |
+
"learning_rate": 1.8145161290322583e-05,
|
| 1294 |
+
"loss": 0.4226,
|
| 1295 |
+
"step": 1670
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 13.46693386773547,
|
| 1299 |
+
"grad_norm": 9.602310180664062,
|
| 1300 |
+
"learning_rate": 1.7921146953405018e-05,
|
| 1301 |
+
"loss": 0.3638,
|
| 1302 |
+
"step": 1680
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 13.547094188376754,
|
| 1306 |
+
"grad_norm": 10.584216117858887,
|
| 1307 |
+
"learning_rate": 1.7697132616487457e-05,
|
| 1308 |
+
"loss": 0.3955,
|
| 1309 |
+
"step": 1690
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 13.627254509018035,
|
| 1313 |
+
"grad_norm": 9.666451454162598,
|
| 1314 |
+
"learning_rate": 1.7473118279569895e-05,
|
| 1315 |
+
"loss": 0.3895,
|
| 1316 |
+
"step": 1700
|
| 1317 |
},
|
| 1318 |
{
|
| 1319 |
+
"epoch": 13.707414829659319,
|
| 1320 |
+
"grad_norm": 14.383480072021484,
|
| 1321 |
+
"learning_rate": 1.724910394265233e-05,
|
| 1322 |
+
"loss": 0.4076,
|
| 1323 |
+
"step": 1710
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 13.7875751503006,
|
| 1327 |
+
"grad_norm": 9.302132606506348,
|
| 1328 |
+
"learning_rate": 1.702508960573477e-05,
|
| 1329 |
+
"loss": 0.4342,
|
| 1330 |
+
"step": 1720
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 13.867735470941884,
|
| 1334 |
+
"grad_norm": 24.193918228149414,
|
| 1335 |
+
"learning_rate": 1.6801075268817204e-05,
|
| 1336 |
+
"loss": 0.4005,
|
| 1337 |
+
"step": 1730
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 13.947895791583166,
|
| 1341 |
+
"grad_norm": 14.272506713867188,
|
| 1342 |
+
"learning_rate": 1.6577060931899643e-05,
|
| 1343 |
+
"loss": 0.4099,
|
| 1344 |
+
"step": 1740
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 13.995991983967937,
|
| 1348 |
+
"eval_accuracy": 0.8624577226606539,
|
| 1349 |
+
"eval_loss": 0.4411380887031555,
|
| 1350 |
+
"eval_runtime": 11.1197,
|
| 1351 |
+
"eval_samples_per_second": 159.537,
|
| 1352 |
+
"eval_steps_per_second": 5.036,
|
| 1353 |
+
"step": 1746
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 14.02805611222445,
|
| 1357 |
+
"grad_norm": 10.529751777648926,
|
| 1358 |
+
"learning_rate": 1.635304659498208e-05,
|
| 1359 |
+
"loss": 0.3849,
|
| 1360 |
+
"step": 1750
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 14.108216432865731,
|
| 1364 |
+
"grad_norm": 9.820696830749512,
|
| 1365 |
+
"learning_rate": 1.6129032258064517e-05,
|
| 1366 |
+
"loss": 0.4455,
|
| 1367 |
+
"step": 1760
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 14.188376753507015,
|
| 1371 |
+
"grad_norm": 8.576085090637207,
|
| 1372 |
+
"learning_rate": 1.5905017921146952e-05,
|
| 1373 |
+
"loss": 0.3852,
|
| 1374 |
+
"step": 1770
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 14.268537074148297,
|
| 1378 |
+
"grad_norm": 11.380485534667969,
|
| 1379 |
+
"learning_rate": 1.568100358422939e-05,
|
| 1380 |
+
"loss": 0.367,
|
| 1381 |
+
"step": 1780
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 14.348697394789578,
|
| 1385 |
+
"grad_norm": 12.20594310760498,
|
| 1386 |
+
"learning_rate": 1.545698924731183e-05,
|
| 1387 |
+
"loss": 0.3974,
|
| 1388 |
+
"step": 1790
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 14.428857715430862,
|
| 1392 |
+
"grad_norm": 11.483406066894531,
|
| 1393 |
+
"learning_rate": 1.5232974910394265e-05,
|
| 1394 |
+
"loss": 0.3949,
|
| 1395 |
+
"step": 1800
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 14.509018036072144,
|
| 1399 |
+
"grad_norm": 9.642448425292969,
|
| 1400 |
+
"learning_rate": 1.5008960573476701e-05,
|
| 1401 |
+
"loss": 0.4097,
|
| 1402 |
+
"step": 1810
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 14.589178356713427,
|
| 1406 |
+
"grad_norm": 10.316274642944336,
|
| 1407 |
+
"learning_rate": 1.4784946236559142e-05,
|
| 1408 |
+
"loss": 0.3393,
|
| 1409 |
+
"step": 1820
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 14.669338677354709,
|
| 1413 |
+
"grad_norm": 10.923069953918457,
|
| 1414 |
+
"learning_rate": 1.4560931899641577e-05,
|
| 1415 |
+
"loss": 0.3562,
|
| 1416 |
+
"step": 1830
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 14.749498997995993,
|
| 1420 |
+
"grad_norm": 9.884988784790039,
|
| 1421 |
+
"learning_rate": 1.4336917562724014e-05,
|
| 1422 |
+
"loss": 0.3581,
|
| 1423 |
+
"step": 1840
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 14.829659318637274,
|
| 1427 |
+
"grad_norm": 8.461724281311035,
|
| 1428 |
+
"learning_rate": 1.4112903225806454e-05,
|
| 1429 |
+
"loss": 0.3775,
|
| 1430 |
+
"step": 1850
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 14.909819639278558,
|
| 1434 |
+
"grad_norm": 8.028816223144531,
|
| 1435 |
+
"learning_rate": 1.388888888888889e-05,
|
| 1436 |
+
"loss": 0.3968,
|
| 1437 |
+
"step": 1860
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 14.98997995991984,
|
| 1441 |
+
"grad_norm": 13.816071510314941,
|
| 1442 |
+
"learning_rate": 1.3664874551971326e-05,
|
| 1443 |
+
"loss": 0.3759,
|
| 1444 |
+
"step": 1870
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 14.997995991983968,
|
| 1448 |
+
"eval_accuracy": 0.8652762119503946,
|
| 1449 |
+
"eval_loss": 0.43173447251319885,
|
| 1450 |
+
"eval_runtime": 13.0155,
|
| 1451 |
+
"eval_samples_per_second": 136.299,
|
| 1452 |
+
"eval_steps_per_second": 4.303,
|
| 1453 |
+
"step": 1871
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 15.070140280561123,
|
| 1457 |
+
"grad_norm": 10.104082107543945,
|
| 1458 |
+
"learning_rate": 1.3440860215053763e-05,
|
| 1459 |
+
"loss": 0.3643,
|
| 1460 |
+
"step": 1880
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 15.150300601202405,
|
| 1464 |
+
"grad_norm": 9.357542991638184,
|
| 1465 |
+
"learning_rate": 1.3216845878136202e-05,
|
| 1466 |
+
"loss": 0.3912,
|
| 1467 |
+
"step": 1890
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 15.230460921843687,
|
| 1471 |
+
"grad_norm": 9.616016387939453,
|
| 1472 |
+
"learning_rate": 1.2992831541218639e-05,
|
| 1473 |
+
"loss": 0.3554,
|
| 1474 |
+
"step": 1900
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 15.31062124248497,
|
| 1478 |
+
"grad_norm": 11.787483215332031,
|
| 1479 |
+
"learning_rate": 1.2768817204301076e-05,
|
| 1480 |
+
"loss": 0.3695,
|
| 1481 |
+
"step": 1910
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 15.390781563126252,
|
| 1485 |
+
"grad_norm": 12.30813980102539,
|
| 1486 |
+
"learning_rate": 1.2544802867383513e-05,
|
| 1487 |
+
"loss": 0.4177,
|
| 1488 |
+
"step": 1920
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 15.470941883767535,
|
| 1492 |
+
"grad_norm": 12.10972785949707,
|
| 1493 |
+
"learning_rate": 1.232078853046595e-05,
|
| 1494 |
+
"loss": 0.3606,
|
| 1495 |
+
"step": 1930
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 15.551102204408817,
|
| 1499 |
+
"grad_norm": 9.871501922607422,
|
| 1500 |
+
"learning_rate": 1.2096774193548388e-05,
|
| 1501 |
+
"loss": 0.3627,
|
| 1502 |
+
"step": 1940
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 15.6312625250501,
|
| 1506 |
+
"grad_norm": 9.91009521484375,
|
| 1507 |
+
"learning_rate": 1.1872759856630825e-05,
|
| 1508 |
+
"loss": 0.363,
|
| 1509 |
+
"step": 1950
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 15.711422845691382,
|
| 1513 |
+
"grad_norm": 9.586908340454102,
|
| 1514 |
+
"learning_rate": 1.1648745519713262e-05,
|
| 1515 |
+
"loss": 0.3761,
|
| 1516 |
+
"step": 1960
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 15.791583166332666,
|
| 1520 |
+
"grad_norm": 13.530887603759766,
|
| 1521 |
+
"learning_rate": 1.14247311827957e-05,
|
| 1522 |
+
"loss": 0.3437,
|
| 1523 |
+
"step": 1970
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 15.871743486973948,
|
| 1527 |
+
"grad_norm": 10.21368408203125,
|
| 1528 |
+
"learning_rate": 1.1200716845878136e-05,
|
| 1529 |
+
"loss": 0.3575,
|
| 1530 |
+
"step": 1980
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 15.951903807615231,
|
| 1534 |
+
"grad_norm": 10.05925178527832,
|
| 1535 |
+
"learning_rate": 1.0976702508960573e-05,
|
| 1536 |
+
"loss": 0.3692,
|
| 1537 |
+
"step": 1990
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 16.0,
|
| 1541 |
+
"eval_accuracy": 0.863021420518602,
|
| 1542 |
+
"eval_loss": 0.43041756749153137,
|
| 1543 |
+
"eval_runtime": 13.2621,
|
| 1544 |
+
"eval_samples_per_second": 133.765,
|
| 1545 |
+
"eval_steps_per_second": 4.223,
|
| 1546 |
+
"step": 1996
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 16.03206412825651,
|
| 1550 |
+
"grad_norm": 8.98168659210205,
|
| 1551 |
+
"learning_rate": 1.0752688172043012e-05,
|
| 1552 |
+
"loss": 0.3292,
|
| 1553 |
+
"step": 2000
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 16.112224448897795,
|
| 1557 |
+
"grad_norm": 13.636373519897461,
|
| 1558 |
+
"learning_rate": 1.0528673835125449e-05,
|
| 1559 |
+
"loss": 0.3546,
|
| 1560 |
+
"step": 2010
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 16.19238476953908,
|
| 1564 |
+
"grad_norm": 8.504558563232422,
|
| 1565 |
+
"learning_rate": 1.0304659498207886e-05,
|
| 1566 |
+
"loss": 0.3346,
|
| 1567 |
+
"step": 2020
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 16.272545090180362,
|
| 1571 |
+
"grad_norm": 12.934548377990723,
|
| 1572 |
+
"learning_rate": 1.0080645161290323e-05,
|
| 1573 |
+
"loss": 0.3453,
|
| 1574 |
+
"step": 2030
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 16.352705410821642,
|
| 1578 |
+
"grad_norm": 10.473663330078125,
|
| 1579 |
+
"learning_rate": 9.856630824372761e-06,
|
| 1580 |
+
"loss": 0.3545,
|
| 1581 |
+
"step": 2040
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 16.432865731462925,
|
| 1585 |
+
"grad_norm": 12.263561248779297,
|
| 1586 |
+
"learning_rate": 9.632616487455196e-06,
|
| 1587 |
+
"loss": 0.3487,
|
| 1588 |
+
"step": 2050
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 16.51302605210421,
|
| 1592 |
+
"grad_norm": 8.574410438537598,
|
| 1593 |
+
"learning_rate": 9.408602150537635e-06,
|
| 1594 |
+
"loss": 0.4023,
|
| 1595 |
+
"step": 2060
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 16.593186372745492,
|
| 1599 |
+
"grad_norm": 20.604251861572266,
|
| 1600 |
+
"learning_rate": 9.184587813620072e-06,
|
| 1601 |
+
"loss": 0.3941,
|
| 1602 |
+
"step": 2070
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 16.673346693386772,
|
| 1606 |
+
"grad_norm": 15.173372268676758,
|
| 1607 |
+
"learning_rate": 8.960573476702509e-06,
|
| 1608 |
+
"loss": 0.3716,
|
| 1609 |
+
"step": 2080
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 16.753507014028056,
|
| 1613 |
+
"grad_norm": 10.56511116027832,
|
| 1614 |
+
"learning_rate": 8.736559139784948e-06,
|
| 1615 |
+
"loss": 0.3873,
|
| 1616 |
+
"step": 2090
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 16.83366733466934,
|
| 1620 |
+
"grad_norm": 12.560215950012207,
|
| 1621 |
+
"learning_rate": 8.512544802867385e-06,
|
| 1622 |
+
"loss": 0.332,
|
| 1623 |
+
"step": 2100
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 16.91382765531062,
|
| 1627 |
+
"grad_norm": 12.049774169921875,
|
| 1628 |
+
"learning_rate": 8.288530465949821e-06,
|
| 1629 |
+
"loss": 0.3394,
|
| 1630 |
+
"step": 2110
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 16.993987975951903,
|
| 1634 |
+
"grad_norm": 12.533961296081543,
|
| 1635 |
+
"learning_rate": 8.064516129032258e-06,
|
| 1636 |
+
"loss": 0.364,
|
| 1637 |
+
"step": 2120
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 16.993987975951903,
|
| 1641 |
+
"eval_accuracy": 0.8664036076662909,
|
| 1642 |
+
"eval_loss": 0.4329654276371002,
|
| 1643 |
+
"eval_runtime": 11.1171,
|
| 1644 |
+
"eval_samples_per_second": 159.574,
|
| 1645 |
+
"eval_steps_per_second": 5.037,
|
| 1646 |
+
"step": 2120
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 17.074148296593187,
|
| 1650 |
+
"grad_norm": 11.729970932006836,
|
| 1651 |
+
"learning_rate": 7.840501792114695e-06,
|
| 1652 |
+
"loss": 0.3719,
|
| 1653 |
+
"step": 2130
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 17.15430861723447,
|
| 1657 |
+
"grad_norm": 8.87394905090332,
|
| 1658 |
+
"learning_rate": 7.616487455197132e-06,
|
| 1659 |
+
"loss": 0.3165,
|
| 1660 |
+
"step": 2140
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 17.23446893787575,
|
| 1664 |
+
"grad_norm": 11.172150611877441,
|
| 1665 |
+
"learning_rate": 7.392473118279571e-06,
|
| 1666 |
+
"loss": 0.371,
|
| 1667 |
+
"step": 2150
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 17.314629258517034,
|
| 1671 |
+
"grad_norm": 17.362049102783203,
|
| 1672 |
+
"learning_rate": 7.168458781362007e-06,
|
| 1673 |
+
"loss": 0.3607,
|
| 1674 |
+
"step": 2160
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 17.394789579158317,
|
| 1678 |
+
"grad_norm": 9.348087310791016,
|
| 1679 |
+
"learning_rate": 6.944444444444445e-06,
|
| 1680 |
+
"loss": 0.3837,
|
| 1681 |
+
"step": 2170
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 17.4749498997996,
|
| 1685 |
+
"grad_norm": 11.758851051330566,
|
| 1686 |
+
"learning_rate": 6.720430107526882e-06,
|
| 1687 |
+
"loss": 0.3847,
|
| 1688 |
+
"step": 2180
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 17.55511022044088,
|
| 1692 |
+
"grad_norm": 12.436318397521973,
|
| 1693 |
+
"learning_rate": 6.4964157706093195e-06,
|
| 1694 |
+
"loss": 0.3541,
|
| 1695 |
+
"step": 2190
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 17.635270541082164,
|
| 1699 |
+
"grad_norm": 8.334653854370117,
|
| 1700 |
+
"learning_rate": 6.2724014336917564e-06,
|
| 1701 |
+
"loss": 0.3638,
|
| 1702 |
+
"step": 2200
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 17.715430861723448,
|
| 1706 |
+
"grad_norm": 13.402957916259766,
|
| 1707 |
+
"learning_rate": 6.048387096774194e-06,
|
| 1708 |
+
"loss": 0.3178,
|
| 1709 |
+
"step": 2210
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 17.79559118236473,
|
| 1713 |
+
"grad_norm": 7.792669296264648,
|
| 1714 |
+
"learning_rate": 5.824372759856631e-06,
|
| 1715 |
+
"loss": 0.3553,
|
| 1716 |
+
"step": 2220
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 17.87575150300601,
|
| 1720 |
+
"grad_norm": 16.891786575317383,
|
| 1721 |
+
"learning_rate": 5.600358422939068e-06,
|
| 1722 |
+
"loss": 0.357,
|
| 1723 |
+
"step": 2230
|
| 1724 |
+
},
|
| 1725 |
+
{
|
| 1726 |
+
"epoch": 17.955911823647295,
|
| 1727 |
+
"grad_norm": 8.470565795898438,
|
| 1728 |
+
"learning_rate": 5.376344086021506e-06,
|
| 1729 |
+
"loss": 0.3636,
|
| 1730 |
+
"step": 2240
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 17.995991983967937,
|
| 1734 |
+
"eval_accuracy": 0.8680947012401353,
|
| 1735 |
+
"eval_loss": 0.4249646067619324,
|
| 1736 |
+
"eval_runtime": 23.3665,
|
| 1737 |
+
"eval_samples_per_second": 75.921,
|
| 1738 |
+
"eval_steps_per_second": 2.397,
|
| 1739 |
+
"step": 2245
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 18.03607214428858,
|
| 1743 |
+
"grad_norm": 8.499645233154297,
|
| 1744 |
+
"learning_rate": 5.152329749103943e-06,
|
| 1745 |
+
"loss": 0.3596,
|
| 1746 |
+
"step": 2250
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 18.11623246492986,
|
| 1750 |
+
"grad_norm": 12.444930076599121,
|
| 1751 |
+
"learning_rate": 4.928315412186381e-06,
|
| 1752 |
+
"loss": 0.2935,
|
| 1753 |
+
"step": 2260
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 18.196392785571142,
|
| 1757 |
+
"grad_norm": 11.180887222290039,
|
| 1758 |
+
"learning_rate": 4.7043010752688175e-06,
|
| 1759 |
+
"loss": 0.3253,
|
| 1760 |
+
"step": 2270
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 18.276553106212425,
|
| 1764 |
+
"grad_norm": 8.559268951416016,
|
| 1765 |
+
"learning_rate": 4.4802867383512545e-06,
|
| 1766 |
+
"loss": 0.3427,
|
| 1767 |
+
"step": 2280
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 18.35671342685371,
|
| 1771 |
+
"grad_norm": 15.954803466796875,
|
| 1772 |
+
"learning_rate": 4.256272401433692e-06,
|
| 1773 |
+
"loss": 0.3657,
|
| 1774 |
+
"step": 2290
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 18.43687374749499,
|
| 1778 |
+
"grad_norm": 12.122089385986328,
|
| 1779 |
+
"learning_rate": 4.032258064516129e-06,
|
| 1780 |
+
"loss": 0.3101,
|
| 1781 |
+
"step": 2300
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 18.517034068136272,
|
| 1785 |
+
"grad_norm": 7.729813575744629,
|
| 1786 |
+
"learning_rate": 3.808243727598566e-06,
|
| 1787 |
+
"loss": 0.3266,
|
| 1788 |
+
"step": 2310
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 18.597194388777556,
|
| 1792 |
+
"grad_norm": 8.754435539245605,
|
| 1793 |
+
"learning_rate": 3.5842293906810035e-06,
|
| 1794 |
+
"loss": 0.3337,
|
| 1795 |
+
"step": 2320
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 18.677354709418836,
|
| 1799 |
+
"grad_norm": 12.22318172454834,
|
| 1800 |
+
"learning_rate": 3.360215053763441e-06,
|
| 1801 |
+
"loss": 0.3262,
|
| 1802 |
+
"step": 2330
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 18.75751503006012,
|
| 1806 |
+
"grad_norm": 11.190741539001465,
|
| 1807 |
+
"learning_rate": 3.1362007168458782e-06,
|
| 1808 |
+
"loss": 0.3537,
|
| 1809 |
+
"step": 2340
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 18.837675350701403,
|
| 1813 |
+
"grad_norm": 12.088068008422852,
|
| 1814 |
+
"learning_rate": 2.9121863799283156e-06,
|
| 1815 |
+
"loss": 0.3034,
|
| 1816 |
+
"step": 2350
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 18.917835671342687,
|
| 1820 |
+
"grad_norm": 8.4877347946167,
|
| 1821 |
+
"learning_rate": 2.688172043010753e-06,
|
| 1822 |
+
"loss": 0.3537,
|
| 1823 |
+
"step": 2360
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 18.997995991983966,
|
| 1827 |
+
"grad_norm": 12.145883560180664,
|
| 1828 |
+
"learning_rate": 2.4641577060931903e-06,
|
| 1829 |
+
"loss": 0.3396,
|
| 1830 |
+
"step": 2370
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 18.997995991983966,
|
| 1834 |
+
"eval_accuracy": 0.8675310033821871,
|
| 1835 |
+
"eval_loss": 0.4275255799293518,
|
| 1836 |
+
"eval_runtime": 20.9813,
|
| 1837 |
+
"eval_samples_per_second": 84.551,
|
| 1838 |
+
"eval_steps_per_second": 2.669,
|
| 1839 |
+
"step": 2370
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 19.07815631262525,
|
| 1843 |
+
"grad_norm": 12.138285636901855,
|
| 1844 |
+
"learning_rate": 2.2401433691756272e-06,
|
| 1845 |
+
"loss": 0.3464,
|
| 1846 |
+
"step": 2380
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 19.158316633266534,
|
| 1850 |
+
"grad_norm": 15.26742172241211,
|
| 1851 |
+
"learning_rate": 2.0161290322580646e-06,
|
| 1852 |
+
"loss": 0.342,
|
| 1853 |
+
"step": 2390
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 19.238476953907817,
|
| 1857 |
+
"grad_norm": 11.72075080871582,
|
| 1858 |
+
"learning_rate": 1.7921146953405017e-06,
|
| 1859 |
+
"loss": 0.3159,
|
| 1860 |
+
"step": 2400
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 19.318637274549097,
|
| 1864 |
+
"grad_norm": 8.406167984008789,
|
| 1865 |
+
"learning_rate": 1.5681003584229391e-06,
|
| 1866 |
+
"loss": 0.2976,
|
| 1867 |
+
"step": 2410
|
| 1868 |
+
},
|
| 1869 |
+
{
|
| 1870 |
+
"epoch": 19.39879759519038,
|
| 1871 |
+
"grad_norm": 9.558813095092773,
|
| 1872 |
+
"learning_rate": 1.3440860215053765e-06,
|
| 1873 |
+
"loss": 0.3713,
|
| 1874 |
+
"step": 2420
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 19.478957915831664,
|
| 1878 |
+
"grad_norm": 12.249772071838379,
|
| 1879 |
+
"learning_rate": 1.1200716845878136e-06,
|
| 1880 |
+
"loss": 0.3353,
|
| 1881 |
+
"step": 2430
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 19.559118236472948,
|
| 1885 |
+
"grad_norm": 14.776098251342773,
|
| 1886 |
+
"learning_rate": 8.960573476702509e-07,
|
| 1887 |
+
"loss": 0.3507,
|
| 1888 |
+
"step": 2440
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 19.639278557114228,
|
| 1892 |
+
"grad_norm": 12.802459716796875,
|
| 1893 |
+
"learning_rate": 6.720430107526882e-07,
|
| 1894 |
+
"loss": 0.3144,
|
| 1895 |
+
"step": 2450
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 19.71943887775551,
|
| 1899 |
+
"grad_norm": 11.719756126403809,
|
| 1900 |
+
"learning_rate": 4.4802867383512544e-07,
|
| 1901 |
+
"loss": 0.3283,
|
| 1902 |
+
"step": 2460
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 19.799599198396795,
|
| 1906 |
+
"grad_norm": 11.714187622070312,
|
| 1907 |
+
"learning_rate": 2.2401433691756272e-07,
|
| 1908 |
+
"loss": 0.3198,
|
| 1909 |
+
"step": 2470
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 19.879759519038075,
|
| 1913 |
+
"grad_norm": 12.341629028320312,
|
| 1914 |
+
"learning_rate": 0.0,
|
| 1915 |
+
"loss": 0.3057,
|
| 1916 |
+
"step": 2480
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 19.879759519038075,
|
| 1920 |
+
"eval_accuracy": 0.8703494926719278,
|
| 1921 |
+
"eval_loss": 0.4261245131492615,
|
| 1922 |
+
"eval_runtime": 20.7326,
|
| 1923 |
+
"eval_samples_per_second": 85.566,
|
| 1924 |
+
"eval_steps_per_second": 2.701,
|
| 1925 |
+
"step": 2480
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 19.879759519038075,
|
| 1929 |
+
"step": 2480,
|
| 1930 |
+
"total_flos": 7.988705158075343e+18,
|
| 1931 |
+
"train_loss": 0.5653726263392356,
|
| 1932 |
+
"train_runtime": 5886.8979,
|
| 1933 |
+
"train_samples_per_second": 54.229,
|
| 1934 |
+
"train_steps_per_second": 0.421
|
| 1935 |
}
|
| 1936 |
],
|
| 1937 |
"logging_steps": 10,
|
| 1938 |
+
"max_steps": 2480,
|
| 1939 |
"num_input_tokens_seen": 0,
|
| 1940 |
+
"num_train_epochs": 20,
|
| 1941 |
"save_steps": 500,
|
| 1942 |
+
"total_flos": 7.988705158075343e+18,
|
| 1943 |
"train_batch_size": 32,
|
| 1944 |
"trial_name": null,
|
| 1945 |
"trial_params": null
|