End of training
Browse files- README.md +3 -3
- all_results.json +10 -10
- eval_results.json +6 -6
- runs/Dec04_00-07-09_DESKTOP-SKBE9FB/events.out.tfevents.1733294025.DESKTOP-SKBE9FB.8600.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +905 -593
README.md
CHANGED
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.8598130841121495
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.4541
|
| 36 |
+
- Accuracy: 0.8598
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 39.
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime": 2.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 2.
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 39.44,
|
| 3 |
+
"eval_accuracy": 0.8598130841121495,
|
| 4 |
+
"eval_loss": 0.45412737131118774,
|
| 5 |
+
"eval_runtime": 2.7464,
|
| 6 |
+
"eval_samples_per_second": 38.961,
|
| 7 |
+
"eval_steps_per_second": 2.549,
|
| 8 |
+
"train_loss": 0.4776322265182223,
|
| 9 |
+
"train_runtime": 1583.2431,
|
| 10 |
+
"train_samples_per_second": 57.047,
|
| 11 |
+
"train_steps_per_second": 0.884
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 39.
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime": 2.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 2.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 39.44,
|
| 3 |
+
"eval_accuracy": 0.8598130841121495,
|
| 4 |
+
"eval_loss": 0.45412737131118774,
|
| 5 |
+
"eval_runtime": 2.7464,
|
| 6 |
+
"eval_samples_per_second": 38.961,
|
| 7 |
+
"eval_steps_per_second": 2.549
|
| 8 |
}
|
runs/Dec04_00-07-09_DESKTOP-SKBE9FB/events.out.tfevents.1733294025.DESKTOP-SKBE9FB.8600.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b65bd365b6b9d50f9c8a7c43e59c3b90db849f85bd82322326bfe77b6d4edb7
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 39.
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 39.44,
|
| 3 |
+
"train_loss": 0.4776322265182223,
|
| 4 |
+
"train_runtime": 1583.2431,
|
| 5 |
+
"train_samples_per_second": 57.047,
|
| 6 |
+
"train_steps_per_second": 0.884
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,917 +1,1229 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "SW2-RHS-DA\\checkpoint-
|
| 4 |
-
"epoch": 39.
|
| 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 |
-
"learning_rate": 3.
|
| 14 |
-
"loss":
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate": 3.
|
| 20 |
-
"loss":
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"
|
| 26 |
-
"
|
| 27 |
-
"eval_runtime": 2.4491,
|
| 28 |
-
"eval_samples_per_second": 43.69,
|
| 29 |
-
"eval_steps_per_second": 2.858,
|
| 30 |
-
"step": 22
|
| 31 |
-
},
|
| 32 |
-
{
|
| 33 |
-
"epoch": 1.33,
|
| 34 |
-
"learning_rate": 3.863636363636364e-05,
|
| 35 |
-
"loss": 0.8548,
|
| 36 |
"step": 30
|
| 37 |
},
|
| 38 |
{
|
| 39 |
-
"epoch":
|
| 40 |
-
"
|
| 41 |
-
"
|
| 42 |
-
"
|
|
|
|
|
|
|
|
|
|
| 43 |
},
|
| 44 |
{
|
| 45 |
-
"epoch":
|
| 46 |
-
"
|
| 47 |
-
"
|
| 48 |
-
"
|
| 49 |
-
"eval_samples_per_second": 42.157,
|
| 50 |
-
"eval_steps_per_second": 2.758,
|
| 51 |
-
"step": 45
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch":
|
| 55 |
-
"learning_rate": 3.
|
| 56 |
-
"loss": 0.
|
| 57 |
"step": 50
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch":
|
| 61 |
-
"learning_rate": 3.
|
| 62 |
-
"loss": 0.
|
| 63 |
"step": 60
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch":
|
| 67 |
-
"
|
| 68 |
-
"
|
| 69 |
-
"eval_runtime": 2.5569,
|
| 70 |
-
"eval_samples_per_second": 41.847,
|
| 71 |
-
"eval_steps_per_second": 2.738,
|
| 72 |
-
"step": 67
|
| 73 |
-
},
|
| 74 |
-
{
|
| 75 |
-
"epoch": 3.11,
|
| 76 |
-
"learning_rate": 3.681818181818182e-05,
|
| 77 |
-
"loss": 0.3143,
|
| 78 |
"step": 70
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch":
|
| 82 |
-
"
|
| 83 |
-
"
|
| 84 |
-
"
|
|
|
|
|
|
|
|
|
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch":
|
| 88 |
-
"learning_rate": 3.
|
| 89 |
-
"loss": 0.
|
| 90 |
-
"step":
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch":
|
| 94 |
-
"
|
| 95 |
-
"
|
| 96 |
-
"eval_runtime": 2.8177,
|
| 97 |
-
"eval_samples_per_second": 37.974,
|
| 98 |
-
"eval_steps_per_second": 2.484,
|
| 99 |
"step": 90
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch":
|
| 103 |
-
"learning_rate": 3.
|
| 104 |
-
"loss": 0.
|
| 105 |
"step": 100
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch":
|
| 109 |
-
"
|
| 110 |
-
"
|
| 111 |
-
"
|
|
|
|
|
|
|
|
|
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch":
|
| 115 |
-
"
|
| 116 |
-
"
|
| 117 |
-
"
|
| 118 |
-
"eval_samples_per_second": 43.152,
|
| 119 |
-
"eval_steps_per_second": 2.823,
|
| 120 |
-
"step": 112
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch":
|
| 124 |
-
"learning_rate": 3.
|
| 125 |
-
"loss": 0.
|
| 126 |
"step": 120
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch":
|
| 130 |
-
"learning_rate": 3.
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 130
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch":
|
| 136 |
-
"
|
| 137 |
-
"
|
| 138 |
-
"eval_runtime": 2.5261,
|
| 139 |
-
"eval_samples_per_second": 42.358,
|
| 140 |
-
"eval_steps_per_second": 2.771,
|
| 141 |
-
"step": 135
|
| 142 |
-
},
|
| 143 |
-
{
|
| 144 |
-
"epoch": 6.22,
|
| 145 |
-
"learning_rate": 3.363636363636364e-05,
|
| 146 |
-
"loss": 0.1082,
|
| 147 |
"step": 140
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch":
|
| 151 |
-
"
|
| 152 |
-
"
|
| 153 |
-
"
|
|
|
|
|
|
|
|
|
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch":
|
| 157 |
-
"
|
| 158 |
-
"
|
| 159 |
-
"
|
| 160 |
-
"eval_samples_per_second": 42.876,
|
| 161 |
-
"eval_steps_per_second": 2.805,
|
| 162 |
-
"step": 157
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"epoch":
|
| 166 |
-
"learning_rate": 3.
|
| 167 |
-
"loss": 0.
|
| 168 |
"step": 160
|
| 169 |
},
|
| 170 |
{
|
| 171 |
-
"epoch":
|
| 172 |
-
"learning_rate": 3.
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 170
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch":
|
| 178 |
-
"
|
| 179 |
-
"
|
| 180 |
-
"
|
|
|
|
|
|
|
|
|
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"epoch":
|
| 184 |
-
"
|
| 185 |
-
"
|
| 186 |
-
"eval_runtime": 2.4861,
|
| 187 |
-
"eval_samples_per_second": 43.04,
|
| 188 |
-
"eval_steps_per_second": 2.816,
|
| 189 |
"step": 180
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"learning_rate": 3.
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 190
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch":
|
| 199 |
-
"learning_rate": 3.
|
| 200 |
-
"loss": 0.
|
| 201 |
"step": 200
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch":
|
| 205 |
-
"
|
| 206 |
-
"
|
| 207 |
-
"eval_runtime": 2.4306,
|
| 208 |
-
"eval_samples_per_second": 44.022,
|
| 209 |
-
"eval_steps_per_second": 2.88,
|
| 210 |
-
"step": 202
|
| 211 |
-
},
|
| 212 |
-
{
|
| 213 |
-
"epoch": 9.33,
|
| 214 |
-
"learning_rate": 3.0454545454545456e-05,
|
| 215 |
-
"loss": 0.0647,
|
| 216 |
"step": 210
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch":
|
| 220 |
-
"
|
| 221 |
-
"
|
| 222 |
-
"
|
|
|
|
|
|
|
|
|
|
| 223 |
},
|
| 224 |
{
|
| 225 |
-
"epoch":
|
| 226 |
-
"
|
| 227 |
-
"
|
| 228 |
-
"
|
| 229 |
-
"eval_samples_per_second": 43.083,
|
| 230 |
-
"eval_steps_per_second": 2.819,
|
| 231 |
-
"step": 225
|
| 232 |
},
|
| 233 |
{
|
| 234 |
-
"epoch":
|
| 235 |
-
"learning_rate":
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 230
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch":
|
| 241 |
-
"learning_rate":
|
| 242 |
-
"loss": 0.
|
| 243 |
"step": 240
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch":
|
| 247 |
-
"eval_accuracy": 0.
|
| 248 |
-
"eval_loss":
|
| 249 |
-
"eval_runtime": 2.
|
| 250 |
-
"eval_samples_per_second": 42.
|
| 251 |
-
"eval_steps_per_second": 2.
|
| 252 |
-
"step":
|
| 253 |
},
|
| 254 |
{
|
| 255 |
-
"epoch":
|
| 256 |
-
"learning_rate":
|
| 257 |
-
"loss": 0.
|
| 258 |
"step": 250
|
| 259 |
},
|
| 260 |
{
|
| 261 |
-
"epoch":
|
| 262 |
-
"learning_rate":
|
| 263 |
-
"loss": 0.
|
| 264 |
"step": 260
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch":
|
| 268 |
-
"learning_rate":
|
| 269 |
-
"loss": 0.
|
| 270 |
"step": 270
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch":
|
| 274 |
-
"
|
| 275 |
-
"
|
| 276 |
-
"eval_runtime": 2.4679,
|
| 277 |
-
"eval_samples_per_second": 43.356,
|
| 278 |
-
"eval_steps_per_second": 2.836,
|
| 279 |
-
"step": 270
|
| 280 |
-
},
|
| 281 |
-
{
|
| 282 |
-
"epoch": 12.44,
|
| 283 |
-
"learning_rate": 2.7272727272727273e-05,
|
| 284 |
-
"loss": 0.0408,
|
| 285 |
"step": 280
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch":
|
| 289 |
-
"
|
| 290 |
-
"
|
| 291 |
-
"
|
|
|
|
|
|
|
|
|
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch":
|
| 295 |
-
"
|
| 296 |
-
"
|
| 297 |
-
"
|
| 298 |
-
"eval_samples_per_second": 43.122,
|
| 299 |
-
"eval_steps_per_second": 2.821,
|
| 300 |
-
"step": 292
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"epoch":
|
| 304 |
-
"learning_rate":
|
| 305 |
-
"loss": 0.
|
| 306 |
"step": 300
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"epoch":
|
| 310 |
-
"learning_rate":
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 310
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch":
|
| 316 |
-
"eval_accuracy": 0.
|
| 317 |
-
"eval_loss":
|
| 318 |
-
"eval_runtime": 2.
|
| 319 |
-
"eval_samples_per_second": 43.
|
| 320 |
-
"eval_steps_per_second": 2.
|
| 321 |
-
"step":
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch":
|
| 325 |
-
"learning_rate":
|
| 326 |
-
"loss": 0.
|
| 327 |
"step": 320
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch":
|
| 331 |
-
"learning_rate":
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 330
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch":
|
| 337 |
-
"
|
| 338 |
-
"
|
| 339 |
-
"eval_runtime": 2.4771,
|
| 340 |
-
"eval_samples_per_second": 43.196,
|
| 341 |
-
"eval_steps_per_second": 2.826,
|
| 342 |
-
"step": 337
|
| 343 |
-
},
|
| 344 |
-
{
|
| 345 |
-
"epoch": 15.11,
|
| 346 |
-
"learning_rate": 2.454545454545455e-05,
|
| 347 |
-
"loss": 0.0273,
|
| 348 |
"step": 340
|
| 349 |
},
|
| 350 |
{
|
| 351 |
-
"epoch":
|
| 352 |
-
"learning_rate":
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 350
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch":
|
| 358 |
-
"
|
| 359 |
-
"
|
| 360 |
-
"
|
|
|
|
|
|
|
|
|
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
-
"
|
| 365 |
-
"
|
| 366 |
-
"eval_runtime": 2.5831,
|
| 367 |
-
"eval_samples_per_second": 41.422,
|
| 368 |
-
"eval_steps_per_second": 2.71,
|
| 369 |
"step": 360
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch":
|
| 373 |
-
"learning_rate": 2.
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 370
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch":
|
| 379 |
-
"learning_rate": 2.
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 380
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"
|
| 386 |
-
"
|
| 387 |
-
"
|
| 388 |
-
"eval_samples_per_second": 44.04,
|
| 389 |
-
"eval_steps_per_second": 2.881,
|
| 390 |
-
"step": 382
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"epoch":
|
| 394 |
-
"
|
| 395 |
-
"
|
|
|
|
|
|
|
|
|
|
| 396 |
"step": 390
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch":
|
| 400 |
-
"learning_rate": 2.
|
| 401 |
-
"loss": 0.
|
| 402 |
"step": 400
|
| 403 |
},
|
| 404 |
{
|
| 405 |
-
"epoch":
|
| 406 |
-
"
|
| 407 |
-
"
|
| 408 |
-
"eval_runtime": 2.5226,
|
| 409 |
-
"eval_samples_per_second": 42.417,
|
| 410 |
-
"eval_steps_per_second": 2.775,
|
| 411 |
-
"step": 405
|
| 412 |
-
},
|
| 413 |
-
{
|
| 414 |
-
"epoch": 18.22,
|
| 415 |
-
"learning_rate": 2.1363636363636365e-05,
|
| 416 |
-
"loss": 0.0419,
|
| 417 |
"step": 410
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch":
|
| 421 |
-
"learning_rate": 2.
|
| 422 |
-
"loss": 0.
|
| 423 |
"step": 420
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch":
|
| 427 |
-
"eval_accuracy": 0.
|
| 428 |
-
"eval_loss":
|
| 429 |
-
"eval_runtime": 2.
|
| 430 |
-
"eval_samples_per_second":
|
| 431 |
-
"eval_steps_per_second": 2.
|
| 432 |
-
"step":
|
| 433 |
},
|
| 434 |
{
|
| 435 |
-
"epoch":
|
| 436 |
-
"learning_rate": 2.
|
| 437 |
-
"loss": 0.
|
| 438 |
"step": 430
|
| 439 |
},
|
| 440 |
{
|
| 441 |
-
"epoch":
|
| 442 |
-
"learning_rate":
|
| 443 |
-
"loss": 0.
|
| 444 |
"step": 440
|
| 445 |
},
|
| 446 |
{
|
| 447 |
-
"epoch":
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss": 0.
|
| 450 |
-
"step": 450
|
| 451 |
-
},
|
| 452 |
-
{
|
| 453 |
-
"epoch": 20.0,
|
| 454 |
-
"eval_accuracy": 0.5887850467289719,
|
| 455 |
-
"eval_loss": 2.5467782020568848,
|
| 456 |
-
"eval_runtime": 2.4846,
|
| 457 |
-
"eval_samples_per_second": 43.065,
|
| 458 |
-
"eval_steps_per_second": 2.817,
|
| 459 |
"step": 450
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch":
|
| 463 |
-
"learning_rate":
|
| 464 |
-
"loss": 0.
|
| 465 |
"step": 460
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch":
|
| 469 |
-
"
|
| 470 |
-
"
|
| 471 |
-
"
|
|
|
|
|
|
|
|
|
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch":
|
| 475 |
-
"
|
| 476 |
-
"
|
| 477 |
-
"
|
| 478 |
-
"eval_samples_per_second": 42.108,
|
| 479 |
-
"eval_steps_per_second": 2.755,
|
| 480 |
-
"step": 472
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"epoch":
|
| 484 |
-
"learning_rate":
|
| 485 |
-
"loss": 0.
|
| 486 |
"step": 480
|
| 487 |
},
|
| 488 |
{
|
| 489 |
-
"epoch":
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss": 0.
|
| 492 |
"step": 490
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch":
|
| 496 |
-
"eval_accuracy": 0.
|
| 497 |
-
"eval_loss":
|
| 498 |
-
"eval_runtime": 2.
|
| 499 |
-
"eval_samples_per_second": 43.
|
| 500 |
-
"eval_steps_per_second": 2.
|
| 501 |
-
"step":
|
| 502 |
},
|
| 503 |
{
|
| 504 |
-
"epoch":
|
| 505 |
-
"learning_rate":
|
| 506 |
-
"loss": 0.
|
| 507 |
"step": 500
|
| 508 |
},
|
| 509 |
{
|
| 510 |
-
"epoch":
|
| 511 |
-
"learning_rate":
|
| 512 |
-
"loss": 0.
|
| 513 |
"step": 510
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch":
|
| 517 |
-
"
|
| 518 |
-
"
|
| 519 |
-
"eval_runtime": 2.5416,
|
| 520 |
-
"eval_samples_per_second": 42.1,
|
| 521 |
-
"eval_steps_per_second": 2.754,
|
| 522 |
-
"step": 517
|
| 523 |
-
},
|
| 524 |
-
{
|
| 525 |
-
"epoch": 23.11,
|
| 526 |
-
"learning_rate": 1.6363636363636366e-05,
|
| 527 |
-
"loss": 0.0293,
|
| 528 |
"step": 520
|
| 529 |
},
|
| 530 |
{
|
| 531 |
-
"epoch":
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 530
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch":
|
| 538 |
-
"
|
| 539 |
-
"
|
| 540 |
-
"
|
|
|
|
|
|
|
|
|
|
| 541 |
},
|
| 542 |
{
|
| 543 |
-
"epoch":
|
| 544 |
-
"
|
| 545 |
-
"
|
| 546 |
-
"eval_runtime": 2.5767,
|
| 547 |
-
"eval_samples_per_second": 41.526,
|
| 548 |
-
"eval_steps_per_second": 2.717,
|
| 549 |
"step": 540
|
| 550 |
},
|
| 551 |
{
|
| 552 |
-
"epoch":
|
| 553 |
-
"learning_rate":
|
| 554 |
-
"loss": 0.
|
| 555 |
"step": 550
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch":
|
| 559 |
-
"learning_rate":
|
| 560 |
-
"loss": 0.
|
| 561 |
"step": 560
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"epoch":
|
| 565 |
-
"eval_accuracy": 0.
|
| 566 |
-
"eval_loss":
|
| 567 |
-
"eval_runtime": 2.
|
| 568 |
-
"eval_samples_per_second":
|
| 569 |
-
"eval_steps_per_second": 2.
|
| 570 |
-
"step":
|
| 571 |
},
|
| 572 |
{
|
| 573 |
-
"epoch":
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 570
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch":
|
| 580 |
-
"learning_rate":
|
| 581 |
-
"loss": 0.
|
| 582 |
"step": 580
|
| 583 |
},
|
| 584 |
{
|
| 585 |
-
"epoch":
|
| 586 |
-
"
|
| 587 |
-
"
|
| 588 |
-
"eval_runtime": 2.5651,
|
| 589 |
-
"eval_samples_per_second": 41.714,
|
| 590 |
-
"eval_steps_per_second": 2.729,
|
| 591 |
-
"step": 585
|
| 592 |
-
},
|
| 593 |
-
{
|
| 594 |
-
"epoch": 26.22,
|
| 595 |
-
"learning_rate": 1.3181818181818183e-05,
|
| 596 |
-
"loss": 0.0185,
|
| 597 |
"step": 590
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch":
|
| 601 |
-
"learning_rate":
|
| 602 |
-
"loss": 0.
|
| 603 |
"step": 600
|
| 604 |
},
|
| 605 |
{
|
| 606 |
-
"epoch":
|
| 607 |
-
"eval_accuracy": 0.
|
| 608 |
-
"eval_loss":
|
| 609 |
-
"eval_runtime": 2.
|
| 610 |
-
"eval_samples_per_second":
|
| 611 |
-
"eval_steps_per_second": 2.
|
| 612 |
-
"step":
|
| 613 |
},
|
| 614 |
{
|
| 615 |
-
"epoch":
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 610
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch":
|
| 622 |
-
"learning_rate":
|
| 623 |
-
"loss": 0.
|
| 624 |
"step": 620
|
| 625 |
},
|
| 626 |
{
|
| 627 |
-
"epoch":
|
| 628 |
-
"learning_rate":
|
| 629 |
-
"loss": 0.
|
| 630 |
"step": 630
|
| 631 |
},
|
| 632 |
{
|
| 633 |
-
"epoch":
|
| 634 |
-
"eval_accuracy": 0.
|
| 635 |
-
"eval_loss":
|
| 636 |
-
"eval_runtime": 2.
|
| 637 |
-
"eval_samples_per_second":
|
| 638 |
-
"eval_steps_per_second": 2.
|
| 639 |
-
"step":
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch":
|
| 643 |
-
"learning_rate":
|
| 644 |
-
"loss": 0.
|
| 645 |
"step": 640
|
| 646 |
},
|
| 647 |
{
|
| 648 |
-
"epoch":
|
| 649 |
-
"learning_rate":
|
| 650 |
-
"loss": 0.
|
| 651 |
"step": 650
|
| 652 |
},
|
| 653 |
{
|
| 654 |
-
"epoch":
|
| 655 |
-
"
|
| 656 |
-
"
|
| 657 |
-
"eval_runtime": 2.4616,
|
| 658 |
-
"eval_samples_per_second": 43.468,
|
| 659 |
-
"eval_steps_per_second": 2.844,
|
| 660 |
-
"step": 652
|
| 661 |
-
},
|
| 662 |
-
{
|
| 663 |
-
"epoch": 29.33,
|
| 664 |
-
"learning_rate": 1e-05,
|
| 665 |
-
"loss": 0.0198,
|
| 666 |
"step": 660
|
| 667 |
},
|
| 668 |
{
|
| 669 |
-
"epoch":
|
| 670 |
-
"learning_rate":
|
| 671 |
-
"loss": 0.
|
| 672 |
"step": 670
|
| 673 |
},
|
| 674 |
{
|
| 675 |
-
"epoch":
|
| 676 |
-
"eval_accuracy": 0.
|
| 677 |
-
"eval_loss":
|
| 678 |
-
"eval_runtime": 2.
|
| 679 |
-
"eval_samples_per_second":
|
| 680 |
-
"eval_steps_per_second": 2.
|
| 681 |
-
"step":
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch":
|
| 685 |
-
"learning_rate":
|
| 686 |
-
"loss": 0.
|
| 687 |
"step": 680
|
| 688 |
},
|
| 689 |
{
|
| 690 |
-
"epoch":
|
| 691 |
-
"learning_rate":
|
| 692 |
-
"loss": 0.
|
| 693 |
"step": 690
|
| 694 |
},
|
| 695 |
{
|
| 696 |
-
"epoch":
|
| 697 |
-
"
|
| 698 |
-
"
|
| 699 |
-
"eval_runtime": 2.5396,
|
| 700 |
-
"eval_samples_per_second": 42.132,
|
| 701 |
-
"eval_steps_per_second": 2.756,
|
| 702 |
-
"step": 697
|
| 703 |
-
},
|
| 704 |
-
{
|
| 705 |
-
"epoch": 31.11,
|
| 706 |
-
"learning_rate": 8.181818181818183e-06,
|
| 707 |
-
"loss": 0.0261,
|
| 708 |
"step": 700
|
| 709 |
},
|
| 710 |
{
|
| 711 |
-
"epoch":
|
| 712 |
-
"learning_rate":
|
| 713 |
-
"loss": 0.
|
| 714 |
"step": 710
|
| 715 |
},
|
| 716 |
{
|
| 717 |
-
"epoch":
|
| 718 |
-
"
|
| 719 |
-
"
|
| 720 |
-
"
|
|
|
|
|
|
|
|
|
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"epoch":
|
| 724 |
-
"
|
| 725 |
-
"
|
| 726 |
-
"eval_runtime": 2.4746,
|
| 727 |
-
"eval_samples_per_second": 43.24,
|
| 728 |
-
"eval_steps_per_second": 2.829,
|
| 729 |
"step": 720
|
| 730 |
},
|
| 731 |
{
|
| 732 |
-
"epoch":
|
| 733 |
-
"learning_rate":
|
| 734 |
-
"loss": 0.
|
| 735 |
"step": 730
|
| 736 |
},
|
| 737 |
{
|
| 738 |
-
"epoch":
|
| 739 |
-
"learning_rate":
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 740
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch":
|
| 745 |
-
"eval_accuracy": 0.
|
| 746 |
-
"eval_loss":
|
| 747 |
-
"eval_runtime": 2.
|
| 748 |
-
"eval_samples_per_second":
|
| 749 |
-
"eval_steps_per_second": 2.
|
| 750 |
-
"step":
|
| 751 |
},
|
| 752 |
{
|
| 753 |
-
"epoch":
|
| 754 |
-
"learning_rate":
|
| 755 |
-
"loss": 0.
|
| 756 |
"step": 750
|
| 757 |
},
|
| 758 |
{
|
| 759 |
-
"epoch":
|
| 760 |
-
"learning_rate":
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 760
|
| 763 |
},
|
| 764 |
{
|
| 765 |
-
"epoch":
|
| 766 |
-
"
|
| 767 |
-
"
|
| 768 |
-
"eval_runtime": 2.4251,
|
| 769 |
-
"eval_samples_per_second": 44.123,
|
| 770 |
-
"eval_steps_per_second": 2.887,
|
| 771 |
-
"step": 765
|
| 772 |
-
},
|
| 773 |
-
{
|
| 774 |
-
"epoch": 34.22,
|
| 775 |
-
"learning_rate": 5e-06,
|
| 776 |
-
"loss": 0.0178,
|
| 777 |
"step": 770
|
| 778 |
},
|
| 779 |
{
|
| 780 |
-
"epoch":
|
| 781 |
-
"learning_rate":
|
| 782 |
-
"loss": 0.
|
| 783 |
"step": 780
|
| 784 |
},
|
| 785 |
{
|
| 786 |
-
"epoch":
|
| 787 |
-
"eval_accuracy": 0.
|
| 788 |
-
"eval_loss":
|
| 789 |
-
"eval_runtime": 2.
|
| 790 |
-
"eval_samples_per_second": 41.
|
| 791 |
-
"eval_steps_per_second": 2.
|
| 792 |
-
"step":
|
| 793 |
},
|
| 794 |
{
|
| 795 |
-
"epoch":
|
| 796 |
-
"learning_rate":
|
| 797 |
-
"loss": 0.
|
| 798 |
"step": 790
|
| 799 |
},
|
| 800 |
{
|
| 801 |
-
"epoch":
|
| 802 |
-
"learning_rate":
|
| 803 |
-
"loss": 0.
|
| 804 |
"step": 800
|
| 805 |
},
|
| 806 |
{
|
| 807 |
-
"epoch":
|
| 808 |
-
"learning_rate":
|
| 809 |
-
"loss": 0.
|
| 810 |
"step": 810
|
| 811 |
},
|
| 812 |
{
|
| 813 |
-
"epoch":
|
| 814 |
-
"eval_accuracy": 0.
|
| 815 |
-
"eval_loss":
|
| 816 |
-
"eval_runtime": 2.
|
| 817 |
-
"eval_samples_per_second": 43.
|
| 818 |
-
"eval_steps_per_second": 2.
|
| 819 |
-
"step":
|
| 820 |
},
|
| 821 |
{
|
| 822 |
-
"epoch":
|
| 823 |
-
"learning_rate":
|
| 824 |
-
"loss": 0.
|
| 825 |
"step": 820
|
| 826 |
},
|
| 827 |
{
|
| 828 |
-
"epoch":
|
| 829 |
-
"learning_rate":
|
| 830 |
-
"loss": 0.
|
| 831 |
"step": 830
|
| 832 |
},
|
| 833 |
{
|
| 834 |
-
"epoch":
|
| 835 |
-
"
|
| 836 |
-
"
|
| 837 |
-
"eval_runtime": 2.46,
|
| 838 |
-
"eval_samples_per_second": 43.495,
|
| 839 |
-
"eval_steps_per_second": 2.845,
|
| 840 |
-
"step": 832
|
| 841 |
-
},
|
| 842 |
-
{
|
| 843 |
-
"epoch": 37.33,
|
| 844 |
-
"learning_rate": 1.8181818181818183e-06,
|
| 845 |
-
"loss": 0.0106,
|
| 846 |
"step": 840
|
| 847 |
},
|
| 848 |
{
|
| 849 |
-
"epoch":
|
| 850 |
-
"learning_rate": 1.
|
| 851 |
-
"loss": 0.
|
| 852 |
"step": 850
|
| 853 |
},
|
| 854 |
{
|
| 855 |
-
"epoch":
|
| 856 |
-
"eval_accuracy": 0.
|
| 857 |
-
"eval_loss":
|
| 858 |
-
"eval_runtime": 2.
|
| 859 |
-
"eval_samples_per_second":
|
| 860 |
-
"eval_steps_per_second": 2.
|
| 861 |
-
"step":
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"epoch":
|
| 865 |
-
"learning_rate":
|
| 866 |
-
"loss": 0.
|
| 867 |
"step": 860
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"epoch":
|
| 871 |
-
"learning_rate":
|
| 872 |
-
"loss": 0.
|
| 873 |
"step": 870
|
| 874 |
},
|
| 875 |
{
|
| 876 |
-
"epoch":
|
| 877 |
-
"
|
| 878 |
-
"
|
| 879 |
-
"
|
| 880 |
-
"eval_samples_per_second": 41.002,
|
| 881 |
-
"eval_steps_per_second": 2.682,
|
| 882 |
-
"step": 877
|
| 883 |
},
|
| 884 |
{
|
| 885 |
-
"epoch":
|
| 886 |
-
"
|
| 887 |
-
"
|
| 888 |
-
"
|
|
|
|
|
|
|
|
|
|
| 889 |
},
|
| 890 |
{
|
| 891 |
-
"epoch":
|
| 892 |
-
"
|
| 893 |
-
"
|
| 894 |
-
"
|
| 895 |
-
|
| 896 |
-
|
| 897 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 898 |
},
|
| 899 |
{
|
| 900 |
-
"epoch": 39
|
| 901 |
-
"
|
| 902 |
-
"
|
| 903 |
-
"
|
| 904 |
-
|
| 905 |
-
|
| 906 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 907 |
}
|
| 908 |
],
|
| 909 |
"logging_steps": 10,
|
| 910 |
-
"max_steps":
|
| 911 |
"num_input_tokens_seen": 0,
|
| 912 |
"num_train_epochs": 40,
|
| 913 |
"save_steps": 500,
|
| 914 |
-
"total_flos":
|
| 915 |
"train_batch_size": 16,
|
| 916 |
"trial_name": null,
|
| 917 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8598130841121495,
|
| 3 |
+
"best_model_checkpoint": "SW2-RHS-DA\\checkpoint-1100",
|
| 4 |
+
"epoch": 39.436619718309856,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1400,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.28,
|
| 13 |
+
"learning_rate": 3.971428571428572e-05,
|
| 14 |
+
"loss": 3.6129,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.56,
|
| 19 |
+
"learning_rate": 3.9428571428571435e-05,
|
| 20 |
+
"loss": 3.1033,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.85,
|
| 25 |
+
"learning_rate": 3.9142857142857145e-05,
|
| 26 |
+
"loss": 2.0538,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.99,
|
| 31 |
+
"eval_accuracy": 0.411214953271028,
|
| 32 |
+
"eval_loss": 1.2866225242614746,
|
| 33 |
+
"eval_runtime": 2.5936,
|
| 34 |
+
"eval_samples_per_second": 41.255,
|
| 35 |
+
"eval_steps_per_second": 2.699,
|
| 36 |
+
"step": 35
|
| 37 |
},
|
| 38 |
{
|
| 39 |
+
"epoch": 1.13,
|
| 40 |
+
"learning_rate": 3.885714285714286e-05,
|
| 41 |
+
"loss": 1.274,
|
| 42 |
+
"step": 40
|
|
|
|
|
|
|
|
|
|
| 43 |
},
|
| 44 |
{
|
| 45 |
+
"epoch": 1.41,
|
| 46 |
+
"learning_rate": 3.857142857142858e-05,
|
| 47 |
+
"loss": 0.7171,
|
| 48 |
"step": 50
|
| 49 |
},
|
| 50 |
{
|
| 51 |
+
"epoch": 1.69,
|
| 52 |
+
"learning_rate": 3.828571428571429e-05,
|
| 53 |
+
"loss": 0.7163,
|
| 54 |
"step": 60
|
| 55 |
},
|
| 56 |
{
|
| 57 |
+
"epoch": 1.97,
|
| 58 |
+
"learning_rate": 3.8e-05,
|
| 59 |
+
"loss": 0.7464,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
"step": 70
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"epoch": 2.0,
|
| 64 |
+
"eval_accuracy": 0.5887850467289719,
|
| 65 |
+
"eval_loss": 0.6779574751853943,
|
| 66 |
+
"eval_runtime": 2.9282,
|
| 67 |
+
"eval_samples_per_second": 36.541,
|
| 68 |
+
"eval_steps_per_second": 2.391,
|
| 69 |
+
"step": 71
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 2.25,
|
| 73 |
+
"learning_rate": 3.771428571428572e-05,
|
| 74 |
+
"loss": 0.7044,
|
| 75 |
+
"step": 80
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 2.54,
|
| 79 |
+
"learning_rate": 3.7428571428571434e-05,
|
| 80 |
+
"loss": 0.6984,
|
|
|
|
|
|
|
|
|
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 2.82,
|
| 85 |
+
"learning_rate": 3.714285714285715e-05,
|
| 86 |
+
"loss": 0.7061,
|
| 87 |
"step": 100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 2.99,
|
| 91 |
+
"eval_accuracy": 0.5887850467289719,
|
| 92 |
+
"eval_loss": 0.6850669980049133,
|
| 93 |
+
"eval_runtime": 2.5306,
|
| 94 |
+
"eval_samples_per_second": 42.282,
|
| 95 |
+
"eval_steps_per_second": 2.766,
|
| 96 |
+
"step": 106
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 3.1,
|
| 100 |
+
"learning_rate": 3.685714285714286e-05,
|
| 101 |
+
"loss": 0.6987,
|
| 102 |
+
"step": 110
|
|
|
|
|
|
|
|
|
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 3.38,
|
| 106 |
+
"learning_rate": 3.6571428571428576e-05,
|
| 107 |
+
"loss": 0.6953,
|
| 108 |
"step": 120
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 3.66,
|
| 112 |
+
"learning_rate": 3.628571428571429e-05,
|
| 113 |
+
"loss": 0.7011,
|
| 114 |
"step": 130
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 3.94,
|
| 118 |
+
"learning_rate": 3.6e-05,
|
| 119 |
+
"loss": 0.6951,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 120 |
"step": 140
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 4.0,
|
| 124 |
+
"eval_accuracy": 0.5887850467289719,
|
| 125 |
+
"eval_loss": 0.6741566061973572,
|
| 126 |
+
"eval_runtime": 2.4201,
|
| 127 |
+
"eval_samples_per_second": 44.212,
|
| 128 |
+
"eval_steps_per_second": 2.892,
|
| 129 |
+
"step": 142
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 4.23,
|
| 133 |
+
"learning_rate": 3.571428571428572e-05,
|
| 134 |
+
"loss": 0.7019,
|
| 135 |
+
"step": 150
|
|
|
|
|
|
|
|
|
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 4.51,
|
| 139 |
+
"learning_rate": 3.542857142857143e-05,
|
| 140 |
+
"loss": 0.6844,
|
| 141 |
"step": 160
|
| 142 |
},
|
| 143 |
{
|
| 144 |
+
"epoch": 4.79,
|
| 145 |
+
"learning_rate": 3.514285714285714e-05,
|
| 146 |
+
"loss": 0.6928,
|
| 147 |
"step": 170
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 4.99,
|
| 151 |
+
"eval_accuracy": 0.4485981308411215,
|
| 152 |
+
"eval_loss": 0.6916905045509338,
|
| 153 |
+
"eval_runtime": 2.487,
|
| 154 |
+
"eval_samples_per_second": 43.024,
|
| 155 |
+
"eval_steps_per_second": 2.815,
|
| 156 |
+
"step": 177
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 5.07,
|
| 160 |
+
"learning_rate": 3.485714285714286e-05,
|
| 161 |
+
"loss": 0.6884,
|
|
|
|
|
|
|
|
|
|
| 162 |
"step": 180
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 5.35,
|
| 166 |
+
"learning_rate": 3.4571428571428574e-05,
|
| 167 |
+
"loss": 0.6884,
|
| 168 |
"step": 190
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 5.63,
|
| 172 |
+
"learning_rate": 3.4285714285714284e-05,
|
| 173 |
+
"loss": 0.6871,
|
| 174 |
"step": 200
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 5.92,
|
| 178 |
+
"learning_rate": 3.4e-05,
|
| 179 |
+
"loss": 0.683,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 180 |
"step": 210
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"epoch": 6.0,
|
| 184 |
+
"eval_accuracy": 0.5794392523364486,
|
| 185 |
+
"eval_loss": 0.6530604362487793,
|
| 186 |
+
"eval_runtime": 2.4047,
|
| 187 |
+
"eval_samples_per_second": 44.497,
|
| 188 |
+
"eval_steps_per_second": 2.911,
|
| 189 |
+
"step": 213
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 6.2,
|
| 193 |
+
"learning_rate": 3.3714285714285716e-05,
|
| 194 |
+
"loss": 0.7098,
|
| 195 |
+
"step": 220
|
|
|
|
|
|
|
|
|
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 6.48,
|
| 199 |
+
"learning_rate": 3.342857142857143e-05,
|
| 200 |
+
"loss": 0.6721,
|
| 201 |
"step": 230
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 6.76,
|
| 205 |
+
"learning_rate": 3.314285714285715e-05,
|
| 206 |
+
"loss": 0.7013,
|
| 207 |
"step": 240
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 6.99,
|
| 211 |
+
"eval_accuracy": 0.4485981308411215,
|
| 212 |
+
"eval_loss": 0.7000461220741272,
|
| 213 |
+
"eval_runtime": 2.4932,
|
| 214 |
+
"eval_samples_per_second": 42.916,
|
| 215 |
+
"eval_steps_per_second": 2.808,
|
| 216 |
+
"step": 248
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 7.04,
|
| 220 |
+
"learning_rate": 3.285714285714286e-05,
|
| 221 |
+
"loss": 0.7038,
|
| 222 |
"step": 250
|
| 223 |
},
|
| 224 |
{
|
| 225 |
+
"epoch": 7.32,
|
| 226 |
+
"learning_rate": 3.257142857142857e-05,
|
| 227 |
+
"loss": 0.6696,
|
| 228 |
"step": 260
|
| 229 |
},
|
| 230 |
{
|
| 231 |
+
"epoch": 7.61,
|
| 232 |
+
"learning_rate": 3.228571428571429e-05,
|
| 233 |
+
"loss": 0.6589,
|
| 234 |
"step": 270
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 7.89,
|
| 238 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 239 |
+
"loss": 0.6921,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 240 |
"step": 280
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 8.0,
|
| 244 |
+
"eval_accuracy": 0.5514018691588785,
|
| 245 |
+
"eval_loss": 0.7518867254257202,
|
| 246 |
+
"eval_runtime": 2.4222,
|
| 247 |
+
"eval_samples_per_second": 44.175,
|
| 248 |
+
"eval_steps_per_second": 2.89,
|
| 249 |
+
"step": 284
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 8.17,
|
| 253 |
+
"learning_rate": 3.1714285714285715e-05,
|
| 254 |
+
"loss": 0.6944,
|
| 255 |
+
"step": 290
|
|
|
|
|
|
|
|
|
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 8.45,
|
| 259 |
+
"learning_rate": 3.142857142857143e-05,
|
| 260 |
+
"loss": 0.6419,
|
| 261 |
"step": 300
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 8.73,
|
| 265 |
+
"learning_rate": 3.114285714285715e-05,
|
| 266 |
+
"loss": 0.6166,
|
| 267 |
"step": 310
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 8.99,
|
| 271 |
+
"eval_accuracy": 0.6822429906542056,
|
| 272 |
+
"eval_loss": 0.5946707129478455,
|
| 273 |
+
"eval_runtime": 2.466,
|
| 274 |
+
"eval_samples_per_second": 43.39,
|
| 275 |
+
"eval_steps_per_second": 2.839,
|
| 276 |
+
"step": 319
|
| 277 |
},
|
| 278 |
{
|
| 279 |
+
"epoch": 9.01,
|
| 280 |
+
"learning_rate": 3.085714285714286e-05,
|
| 281 |
+
"loss": 0.6401,
|
| 282 |
"step": 320
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 9.3,
|
| 286 |
+
"learning_rate": 3.057142857142857e-05,
|
| 287 |
+
"loss": 0.6273,
|
| 288 |
"step": 330
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 9.58,
|
| 292 |
+
"learning_rate": 3.0285714285714288e-05,
|
| 293 |
+
"loss": 0.598,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 294 |
"step": 340
|
| 295 |
},
|
| 296 |
{
|
| 297 |
+
"epoch": 9.86,
|
| 298 |
+
"learning_rate": 3.0000000000000004e-05,
|
| 299 |
+
"loss": 0.6128,
|
| 300 |
"step": 350
|
| 301 |
},
|
| 302 |
{
|
| 303 |
+
"epoch": 10.0,
|
| 304 |
+
"eval_accuracy": 0.7850467289719626,
|
| 305 |
+
"eval_loss": 0.5433729290962219,
|
| 306 |
+
"eval_runtime": 2.6984,
|
| 307 |
+
"eval_samples_per_second": 39.653,
|
| 308 |
+
"eval_steps_per_second": 2.594,
|
| 309 |
+
"step": 355
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 10.14,
|
| 313 |
+
"learning_rate": 2.9714285714285717e-05,
|
| 314 |
+
"loss": 0.6519,
|
|
|
|
|
|
|
|
|
|
| 315 |
"step": 360
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 10.42,
|
| 319 |
+
"learning_rate": 2.9428571428571433e-05,
|
| 320 |
+
"loss": 0.6322,
|
| 321 |
"step": 370
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 10.7,
|
| 325 |
+
"learning_rate": 2.9142857142857146e-05,
|
| 326 |
+
"loss": 0.601,
|
| 327 |
"step": 380
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 10.99,
|
| 331 |
+
"learning_rate": 2.8857142857142858e-05,
|
| 332 |
+
"loss": 0.5737,
|
| 333 |
+
"step": 390
|
|
|
|
|
|
|
|
|
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 10.99,
|
| 337 |
+
"eval_accuracy": 0.7570093457943925,
|
| 338 |
+
"eval_loss": 0.5533133149147034,
|
| 339 |
+
"eval_runtime": 2.4276,
|
| 340 |
+
"eval_samples_per_second": 44.076,
|
| 341 |
+
"eval_steps_per_second": 2.883,
|
| 342 |
"step": 390
|
| 343 |
},
|
| 344 |
{
|
| 345 |
+
"epoch": 11.27,
|
| 346 |
+
"learning_rate": 2.8571428571428574e-05,
|
| 347 |
+
"loss": 0.5554,
|
| 348 |
"step": 400
|
| 349 |
},
|
| 350 |
{
|
| 351 |
+
"epoch": 11.55,
|
| 352 |
+
"learning_rate": 2.828571428571429e-05,
|
| 353 |
+
"loss": 0.587,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 354 |
"step": 410
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 11.83,
|
| 358 |
+
"learning_rate": 2.8e-05,
|
| 359 |
+
"loss": 0.5376,
|
| 360 |
"step": 420
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 12.0,
|
| 364 |
+
"eval_accuracy": 0.7102803738317757,
|
| 365 |
+
"eval_loss": 0.5347476005554199,
|
| 366 |
+
"eval_runtime": 2.4287,
|
| 367 |
+
"eval_samples_per_second": 44.056,
|
| 368 |
+
"eval_steps_per_second": 2.882,
|
| 369 |
+
"step": 426
|
| 370 |
},
|
| 371 |
{
|
| 372 |
+
"epoch": 12.11,
|
| 373 |
+
"learning_rate": 2.7714285714285716e-05,
|
| 374 |
+
"loss": 0.5263,
|
| 375 |
"step": 430
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 12.39,
|
| 379 |
+
"learning_rate": 2.742857142857143e-05,
|
| 380 |
+
"loss": 0.5883,
|
| 381 |
"step": 440
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 12.68,
|
| 385 |
+
"learning_rate": 2.7142857142857148e-05,
|
| 386 |
+
"loss": 0.5797,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 387 |
"step": 450
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 12.96,
|
| 391 |
+
"learning_rate": 2.6857142857142857e-05,
|
| 392 |
+
"loss": 0.5056,
|
| 393 |
"step": 460
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 12.99,
|
| 397 |
+
"eval_accuracy": 0.7663551401869159,
|
| 398 |
+
"eval_loss": 0.4949099123477936,
|
| 399 |
+
"eval_runtime": 2.4981,
|
| 400 |
+
"eval_samples_per_second": 42.833,
|
| 401 |
+
"eval_steps_per_second": 2.802,
|
| 402 |
+
"step": 461
|
| 403 |
},
|
| 404 |
{
|
| 405 |
+
"epoch": 13.24,
|
| 406 |
+
"learning_rate": 2.6571428571428573e-05,
|
| 407 |
+
"loss": 0.5289,
|
| 408 |
+
"step": 470
|
|
|
|
|
|
|
|
|
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 13.52,
|
| 412 |
+
"learning_rate": 2.628571428571429e-05,
|
| 413 |
+
"loss": 0.5917,
|
| 414 |
"step": 480
|
| 415 |
},
|
| 416 |
{
|
| 417 |
+
"epoch": 13.8,
|
| 418 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 419 |
+
"loss": 0.5396,
|
| 420 |
"step": 490
|
| 421 |
},
|
| 422 |
{
|
| 423 |
+
"epoch": 14.0,
|
| 424 |
+
"eval_accuracy": 0.7476635514018691,
|
| 425 |
+
"eval_loss": 0.5150982737541199,
|
| 426 |
+
"eval_runtime": 2.4567,
|
| 427 |
+
"eval_samples_per_second": 43.555,
|
| 428 |
+
"eval_steps_per_second": 2.849,
|
| 429 |
+
"step": 497
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 14.08,
|
| 433 |
+
"learning_rate": 2.5714285714285718e-05,
|
| 434 |
+
"loss": 0.5083,
|
| 435 |
"step": 500
|
| 436 |
},
|
| 437 |
{
|
| 438 |
+
"epoch": 14.37,
|
| 439 |
+
"learning_rate": 2.542857142857143e-05,
|
| 440 |
+
"loss": 0.4903,
|
| 441 |
"step": 510
|
| 442 |
},
|
| 443 |
{
|
| 444 |
+
"epoch": 14.65,
|
| 445 |
+
"learning_rate": 2.5142857142857143e-05,
|
| 446 |
+
"loss": 0.4745,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 447 |
"step": 520
|
| 448 |
},
|
| 449 |
{
|
| 450 |
+
"epoch": 14.93,
|
| 451 |
+
"learning_rate": 2.485714285714286e-05,
|
| 452 |
+
"loss": 0.4826,
|
| 453 |
"step": 530
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 14.99,
|
| 457 |
+
"eval_accuracy": 0.719626168224299,
|
| 458 |
+
"eval_loss": 0.5669169425964355,
|
| 459 |
+
"eval_runtime": 2.4462,
|
| 460 |
+
"eval_samples_per_second": 43.742,
|
| 461 |
+
"eval_steps_per_second": 2.862,
|
| 462 |
+
"step": 532
|
| 463 |
},
|
| 464 |
{
|
| 465 |
+
"epoch": 15.21,
|
| 466 |
+
"learning_rate": 2.4571428571428575e-05,
|
| 467 |
+
"loss": 0.4987,
|
|
|
|
|
|
|
|
|
|
| 468 |
"step": 540
|
| 469 |
},
|
| 470 |
{
|
| 471 |
+
"epoch": 15.49,
|
| 472 |
+
"learning_rate": 2.4285714285714285e-05,
|
| 473 |
+
"loss": 0.4501,
|
| 474 |
"step": 550
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 15.77,
|
| 478 |
+
"learning_rate": 2.4e-05,
|
| 479 |
+
"loss": 0.4269,
|
| 480 |
"step": 560
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"epoch": 16.0,
|
| 484 |
+
"eval_accuracy": 0.7570093457943925,
|
| 485 |
+
"eval_loss": 0.47963711619377136,
|
| 486 |
+
"eval_runtime": 2.4721,
|
| 487 |
+
"eval_samples_per_second": 43.283,
|
| 488 |
+
"eval_steps_per_second": 2.832,
|
| 489 |
+
"step": 568
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 16.06,
|
| 493 |
+
"learning_rate": 2.3714285714285717e-05,
|
| 494 |
+
"loss": 0.4652,
|
| 495 |
"step": 570
|
| 496 |
},
|
| 497 |
{
|
| 498 |
+
"epoch": 16.34,
|
| 499 |
+
"learning_rate": 2.3428571428571433e-05,
|
| 500 |
+
"loss": 0.4804,
|
| 501 |
"step": 580
|
| 502 |
},
|
| 503 |
{
|
| 504 |
+
"epoch": 16.62,
|
| 505 |
+
"learning_rate": 2.3142857142857145e-05,
|
| 506 |
+
"loss": 0.4738,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 507 |
"step": 590
|
| 508 |
},
|
| 509 |
{
|
| 510 |
+
"epoch": 16.9,
|
| 511 |
+
"learning_rate": 2.2857142857142858e-05,
|
| 512 |
+
"loss": 0.5004,
|
| 513 |
"step": 600
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 16.99,
|
| 517 |
+
"eval_accuracy": 0.8037383177570093,
|
| 518 |
+
"eval_loss": 0.4488687813282013,
|
| 519 |
+
"eval_runtime": 2.4276,
|
| 520 |
+
"eval_samples_per_second": 44.077,
|
| 521 |
+
"eval_steps_per_second": 2.884,
|
| 522 |
+
"step": 603
|
| 523 |
},
|
| 524 |
{
|
| 525 |
+
"epoch": 17.18,
|
| 526 |
+
"learning_rate": 2.2571428571428574e-05,
|
| 527 |
+
"loss": 0.424,
|
| 528 |
"step": 610
|
| 529 |
},
|
| 530 |
{
|
| 531 |
+
"epoch": 17.46,
|
| 532 |
+
"learning_rate": 2.2285714285714287e-05,
|
| 533 |
+
"loss": 0.4187,
|
| 534 |
"step": 620
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 17.75,
|
| 538 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 539 |
+
"loss": 0.4116,
|
| 540 |
"step": 630
|
| 541 |
},
|
| 542 |
{
|
| 543 |
+
"epoch": 18.0,
|
| 544 |
+
"eval_accuracy": 0.822429906542056,
|
| 545 |
+
"eval_loss": 0.4361763000488281,
|
| 546 |
+
"eval_runtime": 2.4211,
|
| 547 |
+
"eval_samples_per_second": 44.194,
|
| 548 |
+
"eval_steps_per_second": 2.891,
|
| 549 |
+
"step": 639
|
| 550 |
},
|
| 551 |
{
|
| 552 |
+
"epoch": 18.03,
|
| 553 |
+
"learning_rate": 2.1714285714285715e-05,
|
| 554 |
+
"loss": 0.4238,
|
| 555 |
"step": 640
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 18.31,
|
| 559 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 560 |
+
"loss": 0.3971,
|
| 561 |
"step": 650
|
| 562 |
},
|
| 563 |
{
|
| 564 |
+
"epoch": 18.59,
|
| 565 |
+
"learning_rate": 2.1142857142857144e-05,
|
| 566 |
+
"loss": 0.3699,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 567 |
"step": 660
|
| 568 |
},
|
| 569 |
{
|
| 570 |
+
"epoch": 18.87,
|
| 571 |
+
"learning_rate": 2.085714285714286e-05,
|
| 572 |
+
"loss": 0.3776,
|
| 573 |
"step": 670
|
| 574 |
},
|
| 575 |
{
|
| 576 |
+
"epoch": 18.99,
|
| 577 |
+
"eval_accuracy": 0.7570093457943925,
|
| 578 |
+
"eval_loss": 0.5299957990646362,
|
| 579 |
+
"eval_runtime": 2.4947,
|
| 580 |
+
"eval_samples_per_second": 42.892,
|
| 581 |
+
"eval_steps_per_second": 2.806,
|
| 582 |
+
"step": 674
|
| 583 |
},
|
| 584 |
{
|
| 585 |
+
"epoch": 19.15,
|
| 586 |
+
"learning_rate": 2.057142857142857e-05,
|
| 587 |
+
"loss": 0.43,
|
| 588 |
"step": 680
|
| 589 |
},
|
| 590 |
{
|
| 591 |
+
"epoch": 19.44,
|
| 592 |
+
"learning_rate": 2.0285714285714286e-05,
|
| 593 |
+
"loss": 0.3729,
|
| 594 |
"step": 690
|
| 595 |
},
|
| 596 |
{
|
| 597 |
+
"epoch": 19.72,
|
| 598 |
+
"learning_rate": 2e-05,
|
| 599 |
+
"loss": 0.4071,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 600 |
"step": 700
|
| 601 |
},
|
| 602 |
{
|
| 603 |
+
"epoch": 20.0,
|
| 604 |
+
"learning_rate": 1.9714285714285718e-05,
|
| 605 |
+
"loss": 0.3646,
|
| 606 |
"step": 710
|
| 607 |
},
|
| 608 |
{
|
| 609 |
+
"epoch": 20.0,
|
| 610 |
+
"eval_accuracy": 0.8037383177570093,
|
| 611 |
+
"eval_loss": 0.417547345161438,
|
| 612 |
+
"eval_runtime": 2.539,
|
| 613 |
+
"eval_samples_per_second": 42.142,
|
| 614 |
+
"eval_steps_per_second": 2.757,
|
| 615 |
+
"step": 710
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 20.28,
|
| 619 |
+
"learning_rate": 1.942857142857143e-05,
|
| 620 |
+
"loss": 0.3292,
|
|
|
|
|
|
|
|
|
|
| 621 |
"step": 720
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"epoch": 20.56,
|
| 625 |
+
"learning_rate": 1.9142857142857146e-05,
|
| 626 |
+
"loss": 0.3608,
|
| 627 |
"step": 730
|
| 628 |
},
|
| 629 |
{
|
| 630 |
+
"epoch": 20.85,
|
| 631 |
+
"learning_rate": 1.885714285714286e-05,
|
| 632 |
+
"loss": 0.3683,
|
| 633 |
"step": 740
|
| 634 |
},
|
| 635 |
{
|
| 636 |
+
"epoch": 20.99,
|
| 637 |
+
"eval_accuracy": 0.822429906542056,
|
| 638 |
+
"eval_loss": 0.4699637293815613,
|
| 639 |
+
"eval_runtime": 2.4331,
|
| 640 |
+
"eval_samples_per_second": 43.977,
|
| 641 |
+
"eval_steps_per_second": 2.877,
|
| 642 |
+
"step": 745
|
| 643 |
},
|
| 644 |
{
|
| 645 |
+
"epoch": 21.13,
|
| 646 |
+
"learning_rate": 1.8571428571428575e-05,
|
| 647 |
+
"loss": 0.3365,
|
| 648 |
"step": 750
|
| 649 |
},
|
| 650 |
{
|
| 651 |
+
"epoch": 21.41,
|
| 652 |
+
"learning_rate": 1.8285714285714288e-05,
|
| 653 |
+
"loss": 0.3775,
|
| 654 |
"step": 760
|
| 655 |
},
|
| 656 |
{
|
| 657 |
+
"epoch": 21.69,
|
| 658 |
+
"learning_rate": 1.8e-05,
|
| 659 |
+
"loss": 0.3147,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 660 |
"step": 770
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 21.97,
|
| 664 |
+
"learning_rate": 1.7714285714285717e-05,
|
| 665 |
+
"loss": 0.3277,
|
| 666 |
"step": 780
|
| 667 |
},
|
| 668 |
{
|
| 669 |
+
"epoch": 22.0,
|
| 670 |
+
"eval_accuracy": 0.8130841121495327,
|
| 671 |
+
"eval_loss": 0.470651239156723,
|
| 672 |
+
"eval_runtime": 2.6037,
|
| 673 |
+
"eval_samples_per_second": 41.095,
|
| 674 |
+
"eval_steps_per_second": 2.688,
|
| 675 |
+
"step": 781
|
| 676 |
},
|
| 677 |
{
|
| 678 |
+
"epoch": 22.25,
|
| 679 |
+
"learning_rate": 1.742857142857143e-05,
|
| 680 |
+
"loss": 0.2866,
|
| 681 |
"step": 790
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 22.54,
|
| 685 |
+
"learning_rate": 1.7142857142857142e-05,
|
| 686 |
+
"loss": 0.3304,
|
| 687 |
"step": 800
|
| 688 |
},
|
| 689 |
{
|
| 690 |
+
"epoch": 22.82,
|
| 691 |
+
"learning_rate": 1.6857142857142858e-05,
|
| 692 |
+
"loss": 0.3534,
|
| 693 |
"step": 810
|
| 694 |
},
|
| 695 |
{
|
| 696 |
+
"epoch": 22.99,
|
| 697 |
+
"eval_accuracy": 0.8130841121495327,
|
| 698 |
+
"eval_loss": 0.523983359336853,
|
| 699 |
+
"eval_runtime": 2.439,
|
| 700 |
+
"eval_samples_per_second": 43.871,
|
| 701 |
+
"eval_steps_per_second": 2.87,
|
| 702 |
+
"step": 816
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 23.1,
|
| 706 |
+
"learning_rate": 1.6571428571428574e-05,
|
| 707 |
+
"loss": 0.3282,
|
| 708 |
"step": 820
|
| 709 |
},
|
| 710 |
{
|
| 711 |
+
"epoch": 23.38,
|
| 712 |
+
"learning_rate": 1.6285714285714287e-05,
|
| 713 |
+
"loss": 0.2964,
|
| 714 |
"step": 830
|
| 715 |
},
|
| 716 |
{
|
| 717 |
+
"epoch": 23.66,
|
| 718 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 719 |
+
"loss": 0.3044,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 720 |
"step": 840
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 23.94,
|
| 724 |
+
"learning_rate": 1.5714285714285715e-05,
|
| 725 |
+
"loss": 0.3083,
|
| 726 |
"step": 850
|
| 727 |
},
|
| 728 |
{
|
| 729 |
+
"epoch": 24.0,
|
| 730 |
+
"eval_accuracy": 0.8130841121495327,
|
| 731 |
+
"eval_loss": 0.5011804103851318,
|
| 732 |
+
"eval_runtime": 2.4268,
|
| 733 |
+
"eval_samples_per_second": 44.092,
|
| 734 |
+
"eval_steps_per_second": 2.885,
|
| 735 |
+
"step": 852
|
| 736 |
},
|
| 737 |
{
|
| 738 |
+
"epoch": 24.23,
|
| 739 |
+
"learning_rate": 1.542857142857143e-05,
|
| 740 |
+
"loss": 0.3059,
|
| 741 |
"step": 860
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 24.51,
|
| 745 |
+
"learning_rate": 1.5142857142857144e-05,
|
| 746 |
+
"loss": 0.3426,
|
| 747 |
"step": 870
|
| 748 |
},
|
| 749 |
{
|
| 750 |
+
"epoch": 24.79,
|
| 751 |
+
"learning_rate": 1.4857142857142858e-05,
|
| 752 |
+
"loss": 0.2829,
|
| 753 |
+
"step": 880
|
|
|
|
|
|
|
|
|
|
| 754 |
},
|
| 755 |
{
|
| 756 |
+
"epoch": 24.99,
|
| 757 |
+
"eval_accuracy": 0.8317757009345794,
|
| 758 |
+
"eval_loss": 0.442084401845932,
|
| 759 |
+
"eval_runtime": 2.8308,
|
| 760 |
+
"eval_samples_per_second": 37.798,
|
| 761 |
+
"eval_steps_per_second": 2.473,
|
| 762 |
+
"step": 887
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 25.07,
|
| 766 |
+
"learning_rate": 1.4571428571428573e-05,
|
| 767 |
+
"loss": 0.2844,
|
| 768 |
+
"step": 890
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 25.35,
|
| 772 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 773 |
+
"loss": 0.2989,
|
| 774 |
+
"step": 900
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 25.63,
|
| 778 |
+
"learning_rate": 1.4e-05,
|
| 779 |
+
"loss": 0.2708,
|
| 780 |
+
"step": 910
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 25.92,
|
| 784 |
+
"learning_rate": 1.3714285714285716e-05,
|
| 785 |
+
"loss": 0.2564,
|
| 786 |
+
"step": 920
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 26.0,
|
| 790 |
+
"eval_accuracy": 0.822429906542056,
|
| 791 |
+
"eval_loss": 0.4547804296016693,
|
| 792 |
+
"eval_runtime": 2.4376,
|
| 793 |
+
"eval_samples_per_second": 43.896,
|
| 794 |
+
"eval_steps_per_second": 2.872,
|
| 795 |
+
"step": 923
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 26.2,
|
| 799 |
+
"learning_rate": 1.3428571428571429e-05,
|
| 800 |
+
"loss": 0.2668,
|
| 801 |
+
"step": 930
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 26.48,
|
| 805 |
+
"learning_rate": 1.3142857142857145e-05,
|
| 806 |
+
"loss": 0.266,
|
| 807 |
+
"step": 940
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 26.76,
|
| 811 |
+
"learning_rate": 1.2857142857142859e-05,
|
| 812 |
+
"loss": 0.3136,
|
| 813 |
+
"step": 950
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 26.99,
|
| 817 |
+
"eval_accuracy": 0.8317757009345794,
|
| 818 |
+
"eval_loss": 0.4374113380908966,
|
| 819 |
+
"eval_runtime": 2.5027,
|
| 820 |
+
"eval_samples_per_second": 42.754,
|
| 821 |
+
"eval_steps_per_second": 2.797,
|
| 822 |
+
"step": 958
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 27.04,
|
| 826 |
+
"learning_rate": 1.2571428571428572e-05,
|
| 827 |
+
"loss": 0.2296,
|
| 828 |
+
"step": 960
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 27.32,
|
| 832 |
+
"learning_rate": 1.2285714285714288e-05,
|
| 833 |
+
"loss": 0.3007,
|
| 834 |
+
"step": 970
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 27.61,
|
| 838 |
+
"learning_rate": 1.2e-05,
|
| 839 |
+
"loss": 0.2848,
|
| 840 |
+
"step": 980
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 27.89,
|
| 844 |
+
"learning_rate": 1.1714285714285716e-05,
|
| 845 |
+
"loss": 0.2443,
|
| 846 |
+
"step": 990
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 28.0,
|
| 850 |
+
"eval_accuracy": 0.8130841121495327,
|
| 851 |
+
"eval_loss": 0.5276730060577393,
|
| 852 |
+
"eval_runtime": 2.5662,
|
| 853 |
+
"eval_samples_per_second": 41.696,
|
| 854 |
+
"eval_steps_per_second": 2.728,
|
| 855 |
+
"step": 994
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 28.17,
|
| 859 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 860 |
+
"loss": 0.2526,
|
| 861 |
+
"step": 1000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 28.45,
|
| 865 |
+
"learning_rate": 1.1142857142857143e-05,
|
| 866 |
+
"loss": 0.2539,
|
| 867 |
+
"step": 1010
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 28.73,
|
| 871 |
+
"learning_rate": 1.0857142857142858e-05,
|
| 872 |
+
"loss": 0.258,
|
| 873 |
+
"step": 1020
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 28.99,
|
| 877 |
+
"eval_accuracy": 0.822429906542056,
|
| 878 |
+
"eval_loss": 0.46009838581085205,
|
| 879 |
+
"eval_runtime": 2.4468,
|
| 880 |
+
"eval_samples_per_second": 43.73,
|
| 881 |
+
"eval_steps_per_second": 2.861,
|
| 882 |
+
"step": 1029
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 29.01,
|
| 886 |
+
"learning_rate": 1.0571428571428572e-05,
|
| 887 |
+
"loss": 0.2614,
|
| 888 |
+
"step": 1030
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 29.3,
|
| 892 |
+
"learning_rate": 1.0285714285714285e-05,
|
| 893 |
+
"loss": 0.241,
|
| 894 |
+
"step": 1040
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 29.58,
|
| 898 |
+
"learning_rate": 1e-05,
|
| 899 |
+
"loss": 0.2606,
|
| 900 |
+
"step": 1050
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 29.86,
|
| 904 |
+
"learning_rate": 9.714285714285715e-06,
|
| 905 |
+
"loss": 0.2673,
|
| 906 |
+
"step": 1060
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 30.0,
|
| 910 |
+
"eval_accuracy": 0.8317757009345794,
|
| 911 |
+
"eval_loss": 0.45198604464530945,
|
| 912 |
+
"eval_runtime": 2.4355,
|
| 913 |
+
"eval_samples_per_second": 43.934,
|
| 914 |
+
"eval_steps_per_second": 2.874,
|
| 915 |
+
"step": 1065
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 30.14,
|
| 919 |
+
"learning_rate": 9.42857142857143e-06,
|
| 920 |
+
"loss": 0.2763,
|
| 921 |
+
"step": 1070
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 30.42,
|
| 925 |
+
"learning_rate": 9.142857142857144e-06,
|
| 926 |
+
"loss": 0.2448,
|
| 927 |
+
"step": 1080
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 30.7,
|
| 931 |
+
"learning_rate": 8.857142857142858e-06,
|
| 932 |
+
"loss": 0.2573,
|
| 933 |
+
"step": 1090
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 30.99,
|
| 937 |
+
"learning_rate": 8.571428571428571e-06,
|
| 938 |
+
"loss": 0.2233,
|
| 939 |
+
"step": 1100
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 30.99,
|
| 943 |
+
"eval_accuracy": 0.8598130841121495,
|
| 944 |
+
"eval_loss": 0.45412737131118774,
|
| 945 |
+
"eval_runtime": 2.5403,
|
| 946 |
+
"eval_samples_per_second": 42.12,
|
| 947 |
+
"eval_steps_per_second": 2.756,
|
| 948 |
+
"step": 1100
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 31.27,
|
| 952 |
+
"learning_rate": 8.285714285714287e-06,
|
| 953 |
+
"loss": 0.295,
|
| 954 |
+
"step": 1110
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 31.55,
|
| 958 |
+
"learning_rate": 8.000000000000001e-06,
|
| 959 |
+
"loss": 0.1998,
|
| 960 |
+
"step": 1120
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 31.83,
|
| 964 |
+
"learning_rate": 7.714285714285716e-06,
|
| 965 |
+
"loss": 0.2276,
|
| 966 |
+
"step": 1130
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 32.0,
|
| 970 |
+
"eval_accuracy": 0.8504672897196262,
|
| 971 |
+
"eval_loss": 0.4247356951236725,
|
| 972 |
+
"eval_runtime": 2.4319,
|
| 973 |
+
"eval_samples_per_second": 43.999,
|
| 974 |
+
"eval_steps_per_second": 2.878,
|
| 975 |
+
"step": 1136
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 32.11,
|
| 979 |
+
"learning_rate": 7.428571428571429e-06,
|
| 980 |
+
"loss": 0.2076,
|
| 981 |
+
"step": 1140
|
| 982 |
},
|
| 983 |
{
|
| 984 |
+
"epoch": 32.39,
|
| 985 |
+
"learning_rate": 7.1428571428571436e-06,
|
| 986 |
+
"loss": 0.2375,
|
| 987 |
+
"step": 1150
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 32.68,
|
| 991 |
+
"learning_rate": 6.857142857142858e-06,
|
| 992 |
+
"loss": 0.2523,
|
| 993 |
+
"step": 1160
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 32.96,
|
| 997 |
+
"learning_rate": 6.571428571428572e-06,
|
| 998 |
+
"loss": 0.2653,
|
| 999 |
+
"step": 1170
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 32.99,
|
| 1003 |
+
"eval_accuracy": 0.8504672897196262,
|
| 1004 |
+
"eval_loss": 0.4091378450393677,
|
| 1005 |
+
"eval_runtime": 2.4461,
|
| 1006 |
+
"eval_samples_per_second": 43.744,
|
| 1007 |
+
"eval_steps_per_second": 2.862,
|
| 1008 |
+
"step": 1171
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 33.24,
|
| 1012 |
+
"learning_rate": 6.285714285714286e-06,
|
| 1013 |
+
"loss": 0.2188,
|
| 1014 |
+
"step": 1180
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 33.52,
|
| 1018 |
+
"learning_rate": 6e-06,
|
| 1019 |
+
"loss": 0.2332,
|
| 1020 |
+
"step": 1190
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 33.8,
|
| 1024 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 1025 |
+
"loss": 0.2007,
|
| 1026 |
+
"step": 1200
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 34.0,
|
| 1030 |
+
"eval_accuracy": 0.8504672897196262,
|
| 1031 |
+
"eval_loss": 0.47192811965942383,
|
| 1032 |
+
"eval_runtime": 2.4405,
|
| 1033 |
+
"eval_samples_per_second": 43.843,
|
| 1034 |
+
"eval_steps_per_second": 2.868,
|
| 1035 |
+
"step": 1207
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 34.08,
|
| 1039 |
+
"learning_rate": 5.428571428571429e-06,
|
| 1040 |
+
"loss": 0.2196,
|
| 1041 |
+
"step": 1210
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 34.37,
|
| 1045 |
+
"learning_rate": 5.142857142857142e-06,
|
| 1046 |
+
"loss": 0.2328,
|
| 1047 |
+
"step": 1220
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 34.65,
|
| 1051 |
+
"learning_rate": 4.857142857142858e-06,
|
| 1052 |
+
"loss": 0.1952,
|
| 1053 |
+
"step": 1230
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 34.93,
|
| 1057 |
+
"learning_rate": 4.571428571428572e-06,
|
| 1058 |
+
"loss": 0.2082,
|
| 1059 |
+
"step": 1240
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 34.99,
|
| 1063 |
+
"eval_accuracy": 0.8411214953271028,
|
| 1064 |
+
"eval_loss": 0.4623873233795166,
|
| 1065 |
+
"eval_runtime": 3.047,
|
| 1066 |
+
"eval_samples_per_second": 35.117,
|
| 1067 |
+
"eval_steps_per_second": 2.297,
|
| 1068 |
+
"step": 1242
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 35.21,
|
| 1072 |
+
"learning_rate": 4.2857142857142855e-06,
|
| 1073 |
+
"loss": 0.1957,
|
| 1074 |
+
"step": 1250
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 35.49,
|
| 1078 |
+
"learning_rate": 4.000000000000001e-06,
|
| 1079 |
+
"loss": 0.2021,
|
| 1080 |
+
"step": 1260
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 35.77,
|
| 1084 |
+
"learning_rate": 3.7142857142857146e-06,
|
| 1085 |
+
"loss": 0.1794,
|
| 1086 |
+
"step": 1270
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 36.0,
|
| 1090 |
+
"eval_accuracy": 0.8317757009345794,
|
| 1091 |
+
"eval_loss": 0.48562705516815186,
|
| 1092 |
+
"eval_runtime": 2.5128,
|
| 1093 |
+
"eval_samples_per_second": 42.582,
|
| 1094 |
+
"eval_steps_per_second": 2.786,
|
| 1095 |
+
"step": 1278
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 36.06,
|
| 1099 |
+
"learning_rate": 3.428571428571429e-06,
|
| 1100 |
+
"loss": 0.2378,
|
| 1101 |
+
"step": 1280
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 36.34,
|
| 1105 |
+
"learning_rate": 3.142857142857143e-06,
|
| 1106 |
+
"loss": 0.2363,
|
| 1107 |
+
"step": 1290
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 36.62,
|
| 1111 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 1112 |
+
"loss": 0.2538,
|
| 1113 |
+
"step": 1300
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 36.9,
|
| 1117 |
+
"learning_rate": 2.571428571428571e-06,
|
| 1118 |
+
"loss": 0.1987,
|
| 1119 |
+
"step": 1310
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 36.99,
|
| 1123 |
+
"eval_accuracy": 0.822429906542056,
|
| 1124 |
+
"eval_loss": 0.49035122990608215,
|
| 1125 |
+
"eval_runtime": 2.426,
|
| 1126 |
+
"eval_samples_per_second": 44.106,
|
| 1127 |
+
"eval_steps_per_second": 2.885,
|
| 1128 |
+
"step": 1313
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 37.18,
|
| 1132 |
+
"learning_rate": 2.285714285714286e-06,
|
| 1133 |
+
"loss": 0.2088,
|
| 1134 |
+
"step": 1320
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 37.46,
|
| 1138 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 1139 |
+
"loss": 0.2154,
|
| 1140 |
+
"step": 1330
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 37.75,
|
| 1144 |
+
"learning_rate": 1.7142857142857145e-06,
|
| 1145 |
+
"loss": 0.2066,
|
| 1146 |
+
"step": 1340
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 38.0,
|
| 1150 |
+
"eval_accuracy": 0.8504672897196262,
|
| 1151 |
+
"eval_loss": 0.47410905361175537,
|
| 1152 |
+
"eval_runtime": 2.4985,
|
| 1153 |
+
"eval_samples_per_second": 42.825,
|
| 1154 |
+
"eval_steps_per_second": 2.802,
|
| 1155 |
+
"step": 1349
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 38.03,
|
| 1159 |
+
"learning_rate": 1.4285714285714286e-06,
|
| 1160 |
+
"loss": 0.2381,
|
| 1161 |
+
"step": 1350
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 38.31,
|
| 1165 |
+
"learning_rate": 1.142857142857143e-06,
|
| 1166 |
+
"loss": 0.2242,
|
| 1167 |
+
"step": 1360
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 38.59,
|
| 1171 |
+
"learning_rate": 8.571428571428572e-07,
|
| 1172 |
+
"loss": 0.2374,
|
| 1173 |
+
"step": 1370
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 38.87,
|
| 1177 |
+
"learning_rate": 5.714285714285715e-07,
|
| 1178 |
+
"loss": 0.1972,
|
| 1179 |
+
"step": 1380
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 38.99,
|
| 1183 |
+
"eval_accuracy": 0.8504672897196262,
|
| 1184 |
+
"eval_loss": 0.452989786863327,
|
| 1185 |
+
"eval_runtime": 2.4866,
|
| 1186 |
+
"eval_samples_per_second": 43.031,
|
| 1187 |
+
"eval_steps_per_second": 2.815,
|
| 1188 |
+
"step": 1384
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 39.15,
|
| 1192 |
+
"learning_rate": 2.8571428571428575e-07,
|
| 1193 |
+
"loss": 0.2104,
|
| 1194 |
+
"step": 1390
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 39.44,
|
| 1198 |
+
"learning_rate": 0.0,
|
| 1199 |
+
"loss": 0.2319,
|
| 1200 |
+
"step": 1400
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 39.44,
|
| 1204 |
+
"eval_accuracy": 0.8504672897196262,
|
| 1205 |
+
"eval_loss": 0.45361945033073425,
|
| 1206 |
+
"eval_runtime": 2.5324,
|
| 1207 |
+
"eval_samples_per_second": 42.252,
|
| 1208 |
+
"eval_steps_per_second": 2.764,
|
| 1209 |
+
"step": 1400
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 39.44,
|
| 1213 |
+
"step": 1400,
|
| 1214 |
+
"total_flos": 2.8973120909884785e+18,
|
| 1215 |
+
"train_loss": 0.4776322265182223,
|
| 1216 |
+
"train_runtime": 1583.2431,
|
| 1217 |
+
"train_samples_per_second": 57.047,
|
| 1218 |
+
"train_steps_per_second": 0.884
|
| 1219 |
}
|
| 1220 |
],
|
| 1221 |
"logging_steps": 10,
|
| 1222 |
+
"max_steps": 1400,
|
| 1223 |
"num_input_tokens_seen": 0,
|
| 1224 |
"num_train_epochs": 40,
|
| 1225 |
"save_steps": 500,
|
| 1226 |
+
"total_flos": 2.8973120909884785e+18,
|
| 1227 |
"train_batch_size": 16,
|
| 1228 |
"trial_name": null,
|
| 1229 |
"trial_params": null
|