🍻 cheers
Browse files
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -18,13 +19,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
# vit-lr-cosine-restarts
|
| 20 |
|
| 21 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
-
- Loss: 0.
|
| 24 |
-
- Accuracy: 0.
|
| 25 |
-
- Precision: 0.
|
| 26 |
-
- Recall: 0.
|
| 27 |
-
- F1: 0.
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 19 |
|
| 20 |
# vit-lr-cosine-restarts
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the skin-cancer dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 0.4453
|
| 25 |
+
- Accuracy: 0.8464
|
| 26 |
+
- Precision: 0.8464
|
| 27 |
+
- Recall: 0.8464
|
| 28 |
+
- F1: 0.8438
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
all_results.json
CHANGED
|
@@ -1,16 +1,16 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.61,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_f1": 0.
|
| 5 |
-
"eval_loss": 0.
|
| 6 |
-
"eval_precision": 0.
|
| 7 |
-
"eval_recall": 0.
|
| 8 |
-
"eval_runtime":
|
| 9 |
-
"eval_samples_per_second":
|
| 10 |
-
"eval_steps_per_second": 9.
|
| 11 |
"total_flos": 2.2287694956200755e+18,
|
| 12 |
-
"train_loss": 0.
|
| 13 |
-
"train_runtime":
|
| 14 |
-
"train_samples_per_second":
|
| 15 |
-
"train_steps_per_second":
|
| 16 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.61,
|
| 3 |
+
"eval_accuracy": 0.8463938973647711,
|
| 4 |
+
"eval_f1": 0.8437993661883203,
|
| 5 |
+
"eval_loss": 0.4453237056732178,
|
| 6 |
+
"eval_precision": 0.8463641738950213,
|
| 7 |
+
"eval_recall": 0.8463938973647711,
|
| 8 |
+
"eval_runtime": 38.0107,
|
| 9 |
+
"eval_samples_per_second": 75.873,
|
| 10 |
+
"eval_steps_per_second": 9.497,
|
| 11 |
"total_flos": 2.2287694956200755e+18,
|
| 12 |
+
"train_loss": 0.2811500767639114,
|
| 13 |
+
"train_runtime": 1301.746,
|
| 14 |
+
"train_samples_per_second": 393.932,
|
| 15 |
+
"train_steps_per_second": 24.659
|
| 16 |
}
|
eval_results.json
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.61,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_f1": 0.
|
| 5 |
-
"eval_loss": 0.
|
| 6 |
-
"eval_precision": 0.
|
| 7 |
-
"eval_recall": 0.
|
| 8 |
-
"eval_runtime":
|
| 9 |
-
"eval_samples_per_second":
|
| 10 |
-
"eval_steps_per_second": 9.
|
| 11 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.61,
|
| 3 |
+
"eval_accuracy": 0.8463938973647711,
|
| 4 |
+
"eval_f1": 0.8437993661883203,
|
| 5 |
+
"eval_loss": 0.4453237056732178,
|
| 6 |
+
"eval_precision": 0.8463641738950213,
|
| 7 |
+
"eval_recall": 0.8463938973647711,
|
| 8 |
+
"eval_runtime": 38.0107,
|
| 9 |
+
"eval_samples_per_second": 75.873,
|
| 10 |
+
"eval_steps_per_second": 9.497
|
| 11 |
}
|
runs/Mar19_04-smaller_warmup/events.out.tfevents.1710821117.6492c5bf3fae.6515.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e9fd08d9e079be94e02274b35278b6d5a9bc04ec08c5f79c7f0ddc5729b786d
|
| 3 |
+
size 51502
|
runs/Mar19_04-smaller_warmup/events.out.tfevents.1710822491.6492c5bf3fae.6515.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a0d8b2b7cfec210dbce4ba7c7bc1462a88ec2420f966f908f4a3ad08815e9df
|
| 3 |
+
size 560
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.61,
|
| 3 |
"total_flos": 2.2287694956200755e+18,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.61,
|
| 3 |
"total_flos": 2.2287694956200755e+18,
|
| 4 |
+
"train_loss": 0.2811500767639114,
|
| 5 |
+
"train_runtime": 1301.746,
|
| 6 |
+
"train_samples_per_second": 393.932,
|
| 7 |
+
"train_steps_per_second": 24.659
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
"best_model_checkpoint": "./vit-lr-cosine-restarts/checkpoint-800",
|
| 4 |
"epoch": 5.607476635514018,
|
| 5 |
"eval_steps": 100,
|
|
@@ -10,1488 +10,1488 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.03,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.06,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.09,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.12,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.16,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.19,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.22,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss":
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.25,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.28,
|
| 69 |
-
"grad_norm": 5.
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.31,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.31,
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_f1": 0.
|
| 85 |
-
"eval_loss": 0.
|
| 86 |
-
"eval_precision": 0.
|
| 87 |
-
"eval_recall": 0.
|
| 88 |
-
"eval_runtime":
|
| 89 |
-
"eval_samples_per_second":
|
| 90 |
-
"eval_steps_per_second": 9.
|
| 91 |
"step": 100
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 0.34,
|
| 95 |
-
"grad_norm":
|
| 96 |
-
"learning_rate":
|
| 97 |
-
"loss": 0.
|
| 98 |
"step": 110
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 0.37,
|
| 102 |
-
"grad_norm":
|
| 103 |
-
"learning_rate":
|
| 104 |
-
"loss": 0.
|
| 105 |
"step": 120
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 0.4,
|
| 109 |
-
"grad_norm":
|
| 110 |
-
"learning_rate":
|
| 111 |
-
"loss": 0.
|
| 112 |
"step": 130
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"epoch": 0.44,
|
| 116 |
-
"grad_norm": 3.
|
| 117 |
-
"learning_rate":
|
| 118 |
-
"loss": 0.
|
| 119 |
"step": 140
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 0.47,
|
| 123 |
-
"grad_norm":
|
| 124 |
-
"learning_rate": 9.
|
| 125 |
-
"loss": 0.
|
| 126 |
"step": 150
|
| 127 |
},
|
| 128 |
{
|
| 129 |
"epoch": 0.5,
|
| 130 |
-
"grad_norm":
|
| 131 |
-
"learning_rate": 9.
|
| 132 |
-
"loss": 0.
|
| 133 |
"step": 160
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 0.53,
|
| 137 |
-
"grad_norm": 5.
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 170
|
| 141 |
},
|
| 142 |
{
|
| 143 |
"epoch": 0.56,
|
| 144 |
-
"grad_norm": 6.
|
| 145 |
-
"learning_rate":
|
| 146 |
-
"loss": 0.
|
| 147 |
"step": 180
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 0.59,
|
| 151 |
-
"grad_norm":
|
| 152 |
-
"learning_rate":
|
| 153 |
-
"loss": 0.
|
| 154 |
"step": 190
|
| 155 |
},
|
| 156 |
{
|
| 157 |
"epoch": 0.62,
|
| 158 |
-
"grad_norm":
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss": 0.
|
| 161 |
"step": 200
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 0.62,
|
| 165 |
-
"eval_accuracy": 0.
|
| 166 |
-
"eval_f1": 0.
|
| 167 |
-
"eval_loss": 0.
|
| 168 |
-
"eval_precision": 0.
|
| 169 |
-
"eval_recall": 0.
|
| 170 |
-
"eval_runtime":
|
| 171 |
-
"eval_samples_per_second":
|
| 172 |
-
"eval_steps_per_second": 9.
|
| 173 |
"step": 200
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 0.65,
|
| 177 |
-
"grad_norm":
|
| 178 |
-
"learning_rate":
|
| 179 |
-
"loss": 0.
|
| 180 |
"step": 210
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 0.69,
|
| 184 |
-
"grad_norm":
|
| 185 |
-
"learning_rate":
|
| 186 |
-
"loss": 0.
|
| 187 |
"step": 220
|
| 188 |
},
|
| 189 |
{
|
| 190 |
"epoch": 0.72,
|
| 191 |
-
"grad_norm":
|
| 192 |
-
"learning_rate":
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 230
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 0.75,
|
| 198 |
-
"grad_norm":
|
| 199 |
-
"learning_rate":
|
| 200 |
-
"loss": 0.
|
| 201 |
"step": 240
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 0.78,
|
| 205 |
-
"grad_norm":
|
| 206 |
-
"learning_rate":
|
| 207 |
-
"loss": 0.
|
| 208 |
"step": 250
|
| 209 |
},
|
| 210 |
{
|
| 211 |
"epoch": 0.81,
|
| 212 |
-
"grad_norm":
|
| 213 |
-
"learning_rate":
|
| 214 |
-
"loss": 0.
|
| 215 |
"step": 260
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 0.84,
|
| 219 |
-
"grad_norm":
|
| 220 |
-
"learning_rate":
|
| 221 |
-
"loss": 0.
|
| 222 |
"step": 270
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 0.87,
|
| 226 |
-
"grad_norm":
|
| 227 |
-
"learning_rate":
|
| 228 |
-
"loss": 0.
|
| 229 |
"step": 280
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"epoch": 0.9,
|
| 233 |
-
"grad_norm": 7.
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss": 0.
|
| 236 |
"step": 290
|
| 237 |
},
|
| 238 |
{
|
| 239 |
"epoch": 0.93,
|
| 240 |
-
"grad_norm":
|
| 241 |
-
"learning_rate":
|
| 242 |
-
"loss": 0.
|
| 243 |
"step": 300
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 0.93,
|
| 247 |
-
"eval_accuracy": 0.
|
| 248 |
-
"eval_f1": 0.
|
| 249 |
-
"eval_loss": 0.
|
| 250 |
-
"eval_precision": 0.
|
| 251 |
-
"eval_recall": 0.
|
| 252 |
-
"eval_runtime":
|
| 253 |
-
"eval_samples_per_second":
|
| 254 |
-
"eval_steps_per_second":
|
| 255 |
"step": 300
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.97,
|
| 259 |
-
"grad_norm": 4.
|
| 260 |
-
"learning_rate":
|
| 261 |
-
"loss": 0.
|
| 262 |
"step": 310
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 1.0,
|
| 266 |
-
"grad_norm":
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 320
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 1.03,
|
| 273 |
-
"grad_norm":
|
| 274 |
-
"learning_rate":
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 330
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 1.06,
|
| 280 |
-
"grad_norm": 5.
|
| 281 |
-
"learning_rate":
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 340
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 1.09,
|
| 287 |
-
"grad_norm":
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 350
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 1.12,
|
| 294 |
-
"grad_norm": 3.
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 360
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 1.15,
|
| 301 |
-
"grad_norm":
|
| 302 |
-
"learning_rate":
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 370
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 1.18,
|
| 308 |
-
"grad_norm":
|
| 309 |
-
"learning_rate":
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 380
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 1.21,
|
| 315 |
-
"grad_norm":
|
| 316 |
-
"learning_rate":
|
| 317 |
-
"loss": 0.
|
| 318 |
"step": 390
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 1.25,
|
| 322 |
-
"grad_norm":
|
| 323 |
-
"learning_rate":
|
| 324 |
-
"loss": 0.
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 1.25,
|
| 329 |
-
"eval_accuracy": 0.
|
| 330 |
-
"eval_f1": 0.
|
| 331 |
-
"eval_loss": 0.
|
| 332 |
-
"eval_precision": 0.
|
| 333 |
-
"eval_recall": 0.
|
| 334 |
-
"eval_runtime":
|
| 335 |
-
"eval_samples_per_second":
|
| 336 |
-
"eval_steps_per_second": 9.
|
| 337 |
"step": 400
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 1.28,
|
| 341 |
-
"grad_norm":
|
| 342 |
-
"learning_rate":
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 410
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 1.31,
|
| 348 |
-
"grad_norm":
|
| 349 |
-
"learning_rate":
|
| 350 |
-
"loss": 0.
|
| 351 |
"step": 420
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 1.34,
|
| 355 |
-
"grad_norm":
|
| 356 |
-
"learning_rate":
|
| 357 |
-
"loss": 0.
|
| 358 |
"step": 430
|
| 359 |
},
|
| 360 |
{
|
| 361 |
"epoch": 1.37,
|
| 362 |
-
"grad_norm":
|
| 363 |
-
"learning_rate":
|
| 364 |
-
"loss": 0.
|
| 365 |
"step": 440
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 1.4,
|
| 369 |
-
"grad_norm":
|
| 370 |
-
"learning_rate":
|
| 371 |
-
"loss": 0.
|
| 372 |
"step": 450
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 1.43,
|
| 376 |
-
"grad_norm":
|
| 377 |
-
"learning_rate":
|
| 378 |
-
"loss": 0.
|
| 379 |
"step": 460
|
| 380 |
},
|
| 381 |
{
|
| 382 |
"epoch": 1.46,
|
| 383 |
-
"grad_norm":
|
| 384 |
-
"learning_rate":
|
| 385 |
-
"loss": 0.
|
| 386 |
"step": 470
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 1.5,
|
| 390 |
-
"grad_norm":
|
| 391 |
-
"learning_rate":
|
| 392 |
-
"loss": 0.
|
| 393 |
"step": 480
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 1.53,
|
| 397 |
-
"grad_norm": 7.
|
| 398 |
-
"learning_rate":
|
| 399 |
-
"loss": 0.
|
| 400 |
"step": 490
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 1.56,
|
| 404 |
-
"grad_norm":
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"loss": 0.
|
| 407 |
"step": 500
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 1.56,
|
| 411 |
-
"eval_accuracy": 0.
|
| 412 |
-
"eval_f1": 0.
|
| 413 |
-
"eval_loss": 0.
|
| 414 |
-
"eval_precision": 0.
|
| 415 |
-
"eval_recall": 0.
|
| 416 |
-
"eval_runtime":
|
| 417 |
-
"eval_samples_per_second":
|
| 418 |
-
"eval_steps_per_second": 9.
|
| 419 |
"step": 500
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 1.59,
|
| 423 |
-
"grad_norm":
|
| 424 |
-
"learning_rate":
|
| 425 |
-
"loss": 0.
|
| 426 |
"step": 510
|
| 427 |
},
|
| 428 |
{
|
| 429 |
"epoch": 1.62,
|
| 430 |
-
"grad_norm": 4.
|
| 431 |
-
"learning_rate":
|
| 432 |
-
"loss": 0.
|
| 433 |
"step": 520
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 1.65,
|
| 437 |
-
"grad_norm": 8.
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss": 0.
|
| 440 |
"step": 530
|
| 441 |
},
|
| 442 |
{
|
| 443 |
"epoch": 1.68,
|
| 444 |
-
"grad_norm":
|
| 445 |
-
"learning_rate":
|
| 446 |
-
"loss": 0.
|
| 447 |
"step": 540
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 1.71,
|
| 451 |
-
"grad_norm":
|
| 452 |
-
"learning_rate":
|
| 453 |
-
"loss": 0.
|
| 454 |
"step": 550
|
| 455 |
},
|
| 456 |
{
|
| 457 |
"epoch": 1.74,
|
| 458 |
-
"grad_norm": 4.
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 560
|
| 462 |
},
|
| 463 |
{
|
| 464 |
"epoch": 1.78,
|
| 465 |
-
"grad_norm":
|
| 466 |
-
"learning_rate":
|
| 467 |
-
"loss": 0.
|
| 468 |
"step": 570
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 1.81,
|
| 472 |
-
"grad_norm":
|
| 473 |
-
"learning_rate":
|
| 474 |
-
"loss": 0.
|
| 475 |
"step": 580
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 1.84,
|
| 479 |
-
"grad_norm":
|
| 480 |
-
"learning_rate":
|
| 481 |
-
"loss": 0.
|
| 482 |
"step": 590
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 1.87,
|
| 486 |
-
"grad_norm":
|
| 487 |
-
"learning_rate":
|
| 488 |
-
"loss": 0.
|
| 489 |
"step": 600
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 1.87,
|
| 493 |
-
"eval_accuracy": 0.
|
| 494 |
-
"eval_f1": 0.
|
| 495 |
-
"eval_loss": 0.
|
| 496 |
-
"eval_precision": 0.
|
| 497 |
-
"eval_recall": 0.
|
| 498 |
-
"eval_runtime":
|
| 499 |
-
"eval_samples_per_second":
|
| 500 |
-
"eval_steps_per_second": 9.
|
| 501 |
"step": 600
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 1.9,
|
| 505 |
-
"grad_norm":
|
| 506 |
-
"learning_rate":
|
| 507 |
-
"loss": 0.
|
| 508 |
"step": 610
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 1.93,
|
| 512 |
-
"grad_norm":
|
| 513 |
-
"learning_rate":
|
| 514 |
-
"loss": 0.
|
| 515 |
"step": 620
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 1.96,
|
| 519 |
-
"grad_norm":
|
| 520 |
-
"learning_rate":
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 630
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 1.99,
|
| 526 |
-
"grad_norm":
|
| 527 |
-
"learning_rate":
|
| 528 |
-
"loss": 0.
|
| 529 |
"step": 640
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 2.02,
|
| 533 |
-
"grad_norm":
|
| 534 |
-
"learning_rate":
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 650
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 2.06,
|
| 540 |
-
"grad_norm":
|
| 541 |
-
"learning_rate":
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 660
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 2.09,
|
| 547 |
-
"grad_norm":
|
| 548 |
-
"learning_rate":
|
| 549 |
-
"loss": 0.
|
| 550 |
"step": 670
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 2.12,
|
| 554 |
-
"grad_norm":
|
| 555 |
-
"learning_rate":
|
| 556 |
-
"loss": 0.
|
| 557 |
"step": 680
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 2.15,
|
| 561 |
-
"grad_norm":
|
| 562 |
-
"learning_rate":
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 690
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 2.18,
|
| 568 |
-
"grad_norm":
|
| 569 |
-
"learning_rate":
|
| 570 |
-
"loss": 0.
|
| 571 |
"step": 700
|
| 572 |
},
|
| 573 |
{
|
| 574 |
"epoch": 2.18,
|
| 575 |
-
"eval_accuracy": 0.
|
| 576 |
-
"eval_f1": 0.
|
| 577 |
-
"eval_loss": 0.
|
| 578 |
-
"eval_precision": 0.
|
| 579 |
-
"eval_recall": 0.
|
| 580 |
-
"eval_runtime":
|
| 581 |
-
"eval_samples_per_second":
|
| 582 |
-
"eval_steps_per_second": 9.
|
| 583 |
"step": 700
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 2.21,
|
| 587 |
-
"grad_norm":
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 710
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 2.24,
|
| 594 |
-
"grad_norm":
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 720
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 2.27,
|
| 601 |
-
"grad_norm":
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 730
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 2.31,
|
| 608 |
-
"grad_norm":
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 740
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 2.34,
|
| 615 |
-
"grad_norm": 5.
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 750
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 2.37,
|
| 622 |
-
"grad_norm":
|
| 623 |
-
"learning_rate":
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 760
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 2.4,
|
| 629 |
-
"grad_norm":
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 770
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 2.43,
|
| 636 |
-
"grad_norm":
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 780
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 2.46,
|
| 643 |
-
"grad_norm":
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 790
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 2.49,
|
| 650 |
-
"grad_norm":
|
| 651 |
-
"learning_rate":
|
| 652 |
-
"loss": 0.
|
| 653 |
"step": 800
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 2.49,
|
| 657 |
-
"eval_accuracy": 0.
|
| 658 |
-
"eval_f1": 0.
|
| 659 |
-
"eval_loss": 0.
|
| 660 |
-
"eval_precision": 0.
|
| 661 |
-
"eval_recall": 0.
|
| 662 |
-
"eval_runtime":
|
| 663 |
-
"eval_samples_per_second":
|
| 664 |
-
"eval_steps_per_second": 9.
|
| 665 |
"step": 800
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 2.52,
|
| 669 |
-
"grad_norm":
|
| 670 |
-
"learning_rate":
|
| 671 |
-
"loss": 0.
|
| 672 |
"step": 810
|
| 673 |
},
|
| 674 |
{
|
| 675 |
"epoch": 2.55,
|
| 676 |
-
"grad_norm":
|
| 677 |
-
"learning_rate":
|
| 678 |
-
"loss": 0.
|
| 679 |
"step": 820
|
| 680 |
},
|
| 681 |
{
|
| 682 |
"epoch": 2.59,
|
| 683 |
-
"grad_norm":
|
| 684 |
-
"learning_rate":
|
| 685 |
-
"loss": 0.
|
| 686 |
"step": 830
|
| 687 |
},
|
| 688 |
{
|
| 689 |
"epoch": 2.62,
|
| 690 |
-
"grad_norm":
|
| 691 |
-
"learning_rate":
|
| 692 |
-
"loss": 0.
|
| 693 |
"step": 840
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 2.65,
|
| 697 |
-
"grad_norm":
|
| 698 |
-
"learning_rate":
|
| 699 |
-
"loss": 0.
|
| 700 |
"step": 850
|
| 701 |
},
|
| 702 |
{
|
| 703 |
"epoch": 2.68,
|
| 704 |
-
"grad_norm":
|
| 705 |
-
"learning_rate":
|
| 706 |
-
"loss": 0.
|
| 707 |
"step": 860
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 2.71,
|
| 711 |
-
"grad_norm":
|
| 712 |
-
"learning_rate":
|
| 713 |
-
"loss": 0.
|
| 714 |
"step": 870
|
| 715 |
},
|
| 716 |
{
|
| 717 |
"epoch": 2.74,
|
| 718 |
-
"grad_norm":
|
| 719 |
-
"learning_rate":
|
| 720 |
-
"loss": 0.
|
| 721 |
"step": 880
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 2.77,
|
| 725 |
-
"grad_norm": 2.
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss": 0.
|
| 728 |
"step": 890
|
| 729 |
},
|
| 730 |
{
|
| 731 |
"epoch": 2.8,
|
| 732 |
-
"grad_norm":
|
| 733 |
-
"learning_rate":
|
| 734 |
-
"loss": 0.
|
| 735 |
"step": 900
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 2.8,
|
| 739 |
-
"eval_accuracy": 0.
|
| 740 |
-
"eval_f1": 0.
|
| 741 |
-
"eval_loss": 0.
|
| 742 |
-
"eval_precision": 0.
|
| 743 |
-
"eval_recall": 0.
|
| 744 |
-
"eval_runtime":
|
| 745 |
-
"eval_samples_per_second":
|
| 746 |
-
"eval_steps_per_second": 9.
|
| 747 |
"step": 900
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 2.83,
|
| 751 |
-
"grad_norm":
|
| 752 |
-
"learning_rate":
|
| 753 |
-
"loss": 0.
|
| 754 |
"step": 910
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 2.87,
|
| 758 |
-
"grad_norm":
|
| 759 |
-
"learning_rate":
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 920
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 2.9,
|
| 765 |
-
"grad_norm":
|
| 766 |
-
"learning_rate":
|
| 767 |
-
"loss": 0.
|
| 768 |
"step": 930
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 2.93,
|
| 772 |
-
"grad_norm":
|
| 773 |
-
"learning_rate":
|
| 774 |
-
"loss": 0.
|
| 775 |
"step": 940
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 2.96,
|
| 779 |
-
"grad_norm":
|
| 780 |
-
"learning_rate":
|
| 781 |
-
"loss": 0.
|
| 782 |
"step": 950
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 2.99,
|
| 786 |
-
"grad_norm":
|
| 787 |
-
"learning_rate":
|
| 788 |
-
"loss": 0.
|
| 789 |
"step": 960
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 3.02,
|
| 793 |
-
"grad_norm":
|
| 794 |
-
"learning_rate":
|
| 795 |
-
"loss": 0.
|
| 796 |
"step": 970
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 3.05,
|
| 800 |
-
"grad_norm":
|
| 801 |
-
"learning_rate":
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 980
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 3.08,
|
| 807 |
-
"grad_norm":
|
| 808 |
-
"learning_rate":
|
| 809 |
-
"loss": 0.
|
| 810 |
"step": 990
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 3.12,
|
| 814 |
-
"grad_norm":
|
| 815 |
-
"learning_rate":
|
| 816 |
-
"loss": 0.
|
| 817 |
"step": 1000
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 3.12,
|
| 821 |
-
"eval_accuracy": 0.
|
| 822 |
-
"eval_f1": 0.
|
| 823 |
-
"eval_loss": 0.
|
| 824 |
-
"eval_precision": 0.
|
| 825 |
-
"eval_recall": 0.
|
| 826 |
-
"eval_runtime":
|
| 827 |
-
"eval_samples_per_second":
|
| 828 |
-
"eval_steps_per_second": 9.
|
| 829 |
"step": 1000
|
| 830 |
},
|
| 831 |
{
|
| 832 |
"epoch": 3.15,
|
| 833 |
-
"grad_norm":
|
| 834 |
-
"learning_rate":
|
| 835 |
-
"loss": 0.
|
| 836 |
"step": 1010
|
| 837 |
},
|
| 838 |
{
|
| 839 |
"epoch": 3.18,
|
| 840 |
-
"grad_norm": 4.
|
| 841 |
-
"learning_rate":
|
| 842 |
-
"loss": 0.
|
| 843 |
"step": 1020
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 3.21,
|
| 847 |
-
"grad_norm":
|
| 848 |
-
"learning_rate":
|
| 849 |
-
"loss": 0.
|
| 850 |
"step": 1030
|
| 851 |
},
|
| 852 |
{
|
| 853 |
"epoch": 3.24,
|
| 854 |
-
"grad_norm":
|
| 855 |
-
"learning_rate":
|
| 856 |
-
"loss": 0.
|
| 857 |
"step": 1040
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 3.27,
|
| 861 |
-
"grad_norm":
|
| 862 |
-
"learning_rate":
|
| 863 |
-
"loss": 0.
|
| 864 |
"step": 1050
|
| 865 |
},
|
| 866 |
{
|
| 867 |
"epoch": 3.3,
|
| 868 |
-
"grad_norm":
|
| 869 |
-
"learning_rate":
|
| 870 |
-
"loss": 0.
|
| 871 |
"step": 1060
|
| 872 |
},
|
| 873 |
{
|
| 874 |
"epoch": 3.33,
|
| 875 |
-
"grad_norm":
|
| 876 |
-
"learning_rate":
|
| 877 |
-
"loss": 0.
|
| 878 |
"step": 1070
|
| 879 |
},
|
| 880 |
{
|
| 881 |
"epoch": 3.36,
|
| 882 |
-
"grad_norm": 5.
|
| 883 |
-
"learning_rate":
|
| 884 |
-
"loss": 0.
|
| 885 |
"step": 1080
|
| 886 |
},
|
| 887 |
{
|
| 888 |
"epoch": 3.4,
|
| 889 |
-
"grad_norm":
|
| 890 |
-
"learning_rate":
|
| 891 |
-
"loss": 0.
|
| 892 |
"step": 1090
|
| 893 |
},
|
| 894 |
{
|
| 895 |
"epoch": 3.43,
|
| 896 |
-
"grad_norm":
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss": 0.
|
| 899 |
"step": 1100
|
| 900 |
},
|
| 901 |
{
|
| 902 |
"epoch": 3.43,
|
| 903 |
-
"eval_accuracy": 0.
|
| 904 |
-
"eval_f1": 0.
|
| 905 |
-
"eval_loss": 0.
|
| 906 |
-
"eval_precision": 0.
|
| 907 |
-
"eval_recall": 0.
|
| 908 |
-
"eval_runtime":
|
| 909 |
-
"eval_samples_per_second":
|
| 910 |
-
"eval_steps_per_second": 9.
|
| 911 |
"step": 1100
|
| 912 |
},
|
| 913 |
{
|
| 914 |
"epoch": 3.46,
|
| 915 |
-
"grad_norm":
|
| 916 |
-
"learning_rate":
|
| 917 |
-
"loss": 0.
|
| 918 |
"step": 1110
|
| 919 |
},
|
| 920 |
{
|
| 921 |
"epoch": 3.49,
|
| 922 |
-
"grad_norm":
|
| 923 |
-
"learning_rate":
|
| 924 |
-
"loss": 0.
|
| 925 |
"step": 1120
|
| 926 |
},
|
| 927 |
{
|
| 928 |
"epoch": 3.52,
|
| 929 |
-
"grad_norm":
|
| 930 |
-
"learning_rate":
|
| 931 |
-
"loss": 0.
|
| 932 |
"step": 1130
|
| 933 |
},
|
| 934 |
{
|
| 935 |
"epoch": 3.55,
|
| 936 |
-
"grad_norm":
|
| 937 |
-
"learning_rate":
|
| 938 |
-
"loss": 0.
|
| 939 |
"step": 1140
|
| 940 |
},
|
| 941 |
{
|
| 942 |
"epoch": 3.58,
|
| 943 |
-
"grad_norm":
|
| 944 |
-
"learning_rate":
|
| 945 |
-
"loss": 0.
|
| 946 |
"step": 1150
|
| 947 |
},
|
| 948 |
{
|
| 949 |
"epoch": 3.61,
|
| 950 |
-
"grad_norm":
|
| 951 |
-
"learning_rate":
|
| 952 |
-
"loss": 0.
|
| 953 |
"step": 1160
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 3.64,
|
| 957 |
-
"grad_norm":
|
| 958 |
-
"learning_rate":
|
| 959 |
-
"loss": 0.
|
| 960 |
"step": 1170
|
| 961 |
},
|
| 962 |
{
|
| 963 |
"epoch": 3.68,
|
| 964 |
-
"grad_norm":
|
| 965 |
-
"learning_rate":
|
| 966 |
-
"loss": 0.
|
| 967 |
"step": 1180
|
| 968 |
},
|
| 969 |
{
|
| 970 |
"epoch": 3.71,
|
| 971 |
-
"grad_norm":
|
| 972 |
-
"learning_rate":
|
| 973 |
-
"loss": 0.
|
| 974 |
"step": 1190
|
| 975 |
},
|
| 976 |
{
|
| 977 |
"epoch": 3.74,
|
| 978 |
-
"grad_norm":
|
| 979 |
-
"learning_rate":
|
| 980 |
-
"loss": 0.
|
| 981 |
"step": 1200
|
| 982 |
},
|
| 983 |
{
|
| 984 |
"epoch": 3.74,
|
| 985 |
-
"eval_accuracy": 0.
|
| 986 |
-
"eval_f1": 0.
|
| 987 |
-
"eval_loss": 0.
|
| 988 |
-
"eval_precision": 0.
|
| 989 |
-
"eval_recall": 0.
|
| 990 |
-
"eval_runtime":
|
| 991 |
-
"eval_samples_per_second":
|
| 992 |
-
"eval_steps_per_second":
|
| 993 |
"step": 1200
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 3.77,
|
| 997 |
-
"grad_norm":
|
| 998 |
-
"learning_rate":
|
| 999 |
-
"loss": 0.
|
| 1000 |
"step": 1210
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
"epoch": 3.8,
|
| 1004 |
-
"grad_norm":
|
| 1005 |
-
"learning_rate":
|
| 1006 |
-
"loss": 0.
|
| 1007 |
"step": 1220
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
"epoch": 3.83,
|
| 1011 |
-
"grad_norm":
|
| 1012 |
-
"learning_rate":
|
| 1013 |
-
"loss": 0.
|
| 1014 |
"step": 1230
|
| 1015 |
},
|
| 1016 |
{
|
| 1017 |
"epoch": 3.86,
|
| 1018 |
-
"grad_norm":
|
| 1019 |
-
"learning_rate":
|
| 1020 |
-
"loss": 0.
|
| 1021 |
"step": 1240
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 3.89,
|
| 1025 |
-
"grad_norm":
|
| 1026 |
-
"learning_rate":
|
| 1027 |
-
"loss": 0.
|
| 1028 |
"step": 1250
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
"epoch": 3.93,
|
| 1032 |
-
"grad_norm":
|
| 1033 |
-
"learning_rate":
|
| 1034 |
-
"loss": 0.
|
| 1035 |
"step": 1260
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
"epoch": 3.96,
|
| 1039 |
-
"grad_norm":
|
| 1040 |
-
"learning_rate":
|
| 1041 |
-
"loss": 0.
|
| 1042 |
"step": 1270
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
"epoch": 3.99,
|
| 1046 |
-
"grad_norm":
|
| 1047 |
-
"learning_rate":
|
| 1048 |
-
"loss": 0.
|
| 1049 |
"step": 1280
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
"epoch": 4.02,
|
| 1053 |
-
"grad_norm":
|
| 1054 |
-
"learning_rate":
|
| 1055 |
-
"loss": 0.
|
| 1056 |
"step": 1290
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
"epoch": 4.05,
|
| 1060 |
-
"grad_norm":
|
| 1061 |
-
"learning_rate":
|
| 1062 |
-
"loss": 0.
|
| 1063 |
"step": 1300
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
"epoch": 4.05,
|
| 1067 |
-
"eval_accuracy": 0.
|
| 1068 |
-
"eval_f1": 0.
|
| 1069 |
-
"eval_loss": 0.
|
| 1070 |
-
"eval_precision": 0.
|
| 1071 |
-
"eval_recall": 0.
|
| 1072 |
-
"eval_runtime":
|
| 1073 |
-
"eval_samples_per_second":
|
| 1074 |
-
"eval_steps_per_second":
|
| 1075 |
"step": 1300
|
| 1076 |
},
|
| 1077 |
{
|
| 1078 |
"epoch": 4.08,
|
| 1079 |
-
"grad_norm": 0.
|
| 1080 |
-
"learning_rate":
|
| 1081 |
-
"loss": 0.
|
| 1082 |
"step": 1310
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
"epoch": 4.11,
|
| 1086 |
-
"grad_norm":
|
| 1087 |
-
"learning_rate":
|
| 1088 |
-
"loss": 0.
|
| 1089 |
"step": 1320
|
| 1090 |
},
|
| 1091 |
{
|
| 1092 |
"epoch": 4.14,
|
| 1093 |
-
"grad_norm":
|
| 1094 |
-
"learning_rate":
|
| 1095 |
-
"loss": 0.
|
| 1096 |
"step": 1330
|
| 1097 |
},
|
| 1098 |
{
|
| 1099 |
"epoch": 4.17,
|
| 1100 |
-
"grad_norm":
|
| 1101 |
-
"learning_rate":
|
| 1102 |
-
"loss": 0.
|
| 1103 |
"step": 1340
|
| 1104 |
},
|
| 1105 |
{
|
| 1106 |
"epoch": 4.21,
|
| 1107 |
-
"grad_norm":
|
| 1108 |
-
"learning_rate":
|
| 1109 |
-
"loss": 0.
|
| 1110 |
"step": 1350
|
| 1111 |
},
|
| 1112 |
{
|
| 1113 |
"epoch": 4.24,
|
| 1114 |
-
"grad_norm":
|
| 1115 |
-
"learning_rate":
|
| 1116 |
-
"loss": 0.
|
| 1117 |
"step": 1360
|
| 1118 |
},
|
| 1119 |
{
|
| 1120 |
"epoch": 4.27,
|
| 1121 |
-
"grad_norm":
|
| 1122 |
-
"learning_rate":
|
| 1123 |
-
"loss": 0.
|
| 1124 |
"step": 1370
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
"epoch": 4.3,
|
| 1128 |
-
"grad_norm":
|
| 1129 |
-
"learning_rate":
|
| 1130 |
-
"loss": 0.
|
| 1131 |
"step": 1380
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
"epoch": 4.33,
|
| 1135 |
-
"grad_norm":
|
| 1136 |
-
"learning_rate":
|
| 1137 |
-
"loss": 0.
|
| 1138 |
"step": 1390
|
| 1139 |
},
|
| 1140 |
{
|
| 1141 |
"epoch": 4.36,
|
| 1142 |
-
"grad_norm":
|
| 1143 |
-
"learning_rate": 8.
|
| 1144 |
-
"loss": 0.
|
| 1145 |
"step": 1400
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
"epoch": 4.36,
|
| 1149 |
-
"eval_accuracy": 0.
|
| 1150 |
-
"eval_f1": 0.
|
| 1151 |
-
"eval_loss": 0.
|
| 1152 |
-
"eval_precision": 0.
|
| 1153 |
-
"eval_recall": 0.
|
| 1154 |
-
"eval_runtime":
|
| 1155 |
-
"eval_samples_per_second":
|
| 1156 |
-
"eval_steps_per_second": 9.
|
| 1157 |
"step": 1400
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 4.39,
|
| 1161 |
-
"grad_norm": 0.
|
| 1162 |
-
"learning_rate": 8.
|
| 1163 |
-
"loss": 0.
|
| 1164 |
"step": 1410
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
"epoch": 4.42,
|
| 1168 |
-
"grad_norm":
|
| 1169 |
-
"learning_rate": 8.
|
| 1170 |
-
"loss": 0.
|
| 1171 |
"step": 1420
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 4.45,
|
| 1175 |
-
"grad_norm":
|
| 1176 |
-
"learning_rate": 8.
|
| 1177 |
-
"loss": 0.
|
| 1178 |
"step": 1430
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 4.49,
|
| 1182 |
-
"grad_norm":
|
| 1183 |
-
"learning_rate": 8.
|
| 1184 |
-
"loss": 0.
|
| 1185 |
"step": 1440
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 4.52,
|
| 1189 |
-
"grad_norm":
|
| 1190 |
-
"learning_rate":
|
| 1191 |
-
"loss": 0.
|
| 1192 |
"step": 1450
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
"epoch": 4.55,
|
| 1196 |
-
"grad_norm":
|
| 1197 |
-
"learning_rate":
|
| 1198 |
-
"loss": 0.
|
| 1199 |
"step": 1460
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 4.58,
|
| 1203 |
-
"grad_norm":
|
| 1204 |
-
"learning_rate":
|
| 1205 |
-
"loss": 0.
|
| 1206 |
"step": 1470
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
"epoch": 4.61,
|
| 1210 |
-
"grad_norm":
|
| 1211 |
-
"learning_rate":
|
| 1212 |
-
"loss": 0.
|
| 1213 |
"step": 1480
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 4.64,
|
| 1217 |
-
"grad_norm":
|
| 1218 |
-
"learning_rate":
|
| 1219 |
-
"loss": 0.
|
| 1220 |
"step": 1490
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
"epoch": 4.67,
|
| 1224 |
-
"grad_norm":
|
| 1225 |
-
"learning_rate":
|
| 1226 |
-
"loss": 0.
|
| 1227 |
"step": 1500
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 4.67,
|
| 1231 |
-
"eval_accuracy": 0.
|
| 1232 |
-
"eval_f1": 0.
|
| 1233 |
-
"eval_loss": 0.
|
| 1234 |
-
"eval_precision": 0.
|
| 1235 |
-
"eval_recall": 0.
|
| 1236 |
-
"eval_runtime":
|
| 1237 |
-
"eval_samples_per_second":
|
| 1238 |
-
"eval_steps_per_second": 9.
|
| 1239 |
"step": 1500
|
| 1240 |
},
|
| 1241 |
{
|
| 1242 |
"epoch": 4.7,
|
| 1243 |
-
"grad_norm":
|
| 1244 |
-
"learning_rate":
|
| 1245 |
-
"loss": 0.
|
| 1246 |
"step": 1510
|
| 1247 |
},
|
| 1248 |
{
|
| 1249 |
"epoch": 4.74,
|
| 1250 |
-
"grad_norm":
|
| 1251 |
-
"learning_rate":
|
| 1252 |
-
"loss": 0.
|
| 1253 |
"step": 1520
|
| 1254 |
},
|
| 1255 |
{
|
| 1256 |
"epoch": 4.77,
|
| 1257 |
-
"grad_norm":
|
| 1258 |
-
"learning_rate":
|
| 1259 |
-
"loss": 0.
|
| 1260 |
"step": 1530
|
| 1261 |
},
|
| 1262 |
{
|
| 1263 |
"epoch": 4.8,
|
| 1264 |
-
"grad_norm":
|
| 1265 |
-
"learning_rate":
|
| 1266 |
-
"loss": 0.
|
| 1267 |
"step": 1540
|
| 1268 |
},
|
| 1269 |
{
|
| 1270 |
"epoch": 4.83,
|
| 1271 |
-
"grad_norm":
|
| 1272 |
-
"learning_rate":
|
| 1273 |
-
"loss": 0.
|
| 1274 |
"step": 1550
|
| 1275 |
},
|
| 1276 |
{
|
| 1277 |
"epoch": 4.86,
|
| 1278 |
-
"grad_norm": 5.
|
| 1279 |
-
"learning_rate":
|
| 1280 |
-
"loss": 0.
|
| 1281 |
"step": 1560
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
"epoch": 4.89,
|
| 1285 |
-
"grad_norm":
|
| 1286 |
-
"learning_rate":
|
| 1287 |
-
"loss": 0.
|
| 1288 |
"step": 1570
|
| 1289 |
},
|
| 1290 |
{
|
| 1291 |
"epoch": 4.92,
|
| 1292 |
-
"grad_norm":
|
| 1293 |
-
"learning_rate":
|
| 1294 |
-
"loss": 0.
|
| 1295 |
"step": 1580
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
"epoch": 4.95,
|
| 1299 |
-
"grad_norm":
|
| 1300 |
-
"learning_rate":
|
| 1301 |
-
"loss": 0.
|
| 1302 |
"step": 1590
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
"epoch": 4.98,
|
| 1306 |
-
"grad_norm":
|
| 1307 |
-
"learning_rate":
|
| 1308 |
-
"loss": 0.
|
| 1309 |
"step": 1600
|
| 1310 |
},
|
| 1311 |
{
|
| 1312 |
"epoch": 4.98,
|
| 1313 |
-
"eval_accuracy": 0.
|
| 1314 |
-
"eval_f1": 0.
|
| 1315 |
-
"eval_loss": 0.
|
| 1316 |
-
"eval_precision": 0.
|
| 1317 |
-
"eval_recall": 0.
|
| 1318 |
-
"eval_runtime":
|
| 1319 |
-
"eval_samples_per_second":
|
| 1320 |
-
"eval_steps_per_second": 9.
|
| 1321 |
"step": 1600
|
| 1322 |
},
|
| 1323 |
{
|
| 1324 |
"epoch": 5.02,
|
| 1325 |
-
"grad_norm":
|
| 1326 |
-
"learning_rate":
|
| 1327 |
-
"loss": 0.
|
| 1328 |
"step": 1610
|
| 1329 |
},
|
| 1330 |
{
|
| 1331 |
"epoch": 5.05,
|
| 1332 |
-
"grad_norm":
|
| 1333 |
-
"learning_rate":
|
| 1334 |
-
"loss": 0.
|
| 1335 |
"step": 1620
|
| 1336 |
},
|
| 1337 |
{
|
| 1338 |
"epoch": 5.08,
|
| 1339 |
-
"grad_norm":
|
| 1340 |
-
"learning_rate":
|
| 1341 |
-
"loss": 0.
|
| 1342 |
"step": 1630
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
"epoch": 5.11,
|
| 1346 |
-
"grad_norm": 0.
|
| 1347 |
-
"learning_rate":
|
| 1348 |
-
"loss": 0.
|
| 1349 |
"step": 1640
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
"epoch": 5.14,
|
| 1353 |
-
"grad_norm":
|
| 1354 |
-
"learning_rate":
|
| 1355 |
-
"loss": 0.
|
| 1356 |
"step": 1650
|
| 1357 |
},
|
| 1358 |
{
|
| 1359 |
"epoch": 5.17,
|
| 1360 |
-
"grad_norm": 0.
|
| 1361 |
-
"learning_rate":
|
| 1362 |
-
"loss": 0.
|
| 1363 |
"step": 1660
|
| 1364 |
},
|
| 1365 |
{
|
| 1366 |
"epoch": 5.2,
|
| 1367 |
-
"grad_norm": 0.
|
| 1368 |
-
"learning_rate":
|
| 1369 |
-
"loss": 0.
|
| 1370 |
"step": 1670
|
| 1371 |
},
|
| 1372 |
{
|
| 1373 |
"epoch": 5.23,
|
| 1374 |
-
"grad_norm": 0.
|
| 1375 |
-
"learning_rate":
|
| 1376 |
-
"loss": 0.
|
| 1377 |
"step": 1680
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
"epoch": 5.26,
|
| 1381 |
-
"grad_norm":
|
| 1382 |
-
"learning_rate":
|
| 1383 |
-
"loss": 0.
|
| 1384 |
"step": 1690
|
| 1385 |
},
|
| 1386 |
{
|
| 1387 |
"epoch": 5.3,
|
| 1388 |
-
"grad_norm":
|
| 1389 |
-
"learning_rate":
|
| 1390 |
-
"loss": 0.
|
| 1391 |
"step": 1700
|
| 1392 |
},
|
| 1393 |
{
|
| 1394 |
"epoch": 5.3,
|
| 1395 |
-
"eval_accuracy": 0.
|
| 1396 |
-
"eval_f1": 0.
|
| 1397 |
-
"eval_loss": 0.
|
| 1398 |
-
"eval_precision": 0.
|
| 1399 |
-
"eval_recall": 0.
|
| 1400 |
-
"eval_runtime":
|
| 1401 |
-
"eval_samples_per_second":
|
| 1402 |
-
"eval_steps_per_second": 9.
|
| 1403 |
"step": 1700
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
"epoch": 5.33,
|
| 1407 |
-
"grad_norm": 0.
|
| 1408 |
-
"learning_rate":
|
| 1409 |
-
"loss": 0.
|
| 1410 |
"step": 1710
|
| 1411 |
},
|
| 1412 |
{
|
| 1413 |
"epoch": 5.36,
|
| 1414 |
-
"grad_norm":
|
| 1415 |
-
"learning_rate":
|
| 1416 |
-
"loss": 0.
|
| 1417 |
"step": 1720
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
"epoch": 5.39,
|
| 1421 |
-
"grad_norm":
|
| 1422 |
-
"learning_rate":
|
| 1423 |
-
"loss": 0.
|
| 1424 |
"step": 1730
|
| 1425 |
},
|
| 1426 |
{
|
| 1427 |
"epoch": 5.42,
|
| 1428 |
-
"grad_norm":
|
| 1429 |
-
"learning_rate":
|
| 1430 |
-
"loss": 0.
|
| 1431 |
"step": 1740
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
"epoch": 5.45,
|
| 1435 |
-
"grad_norm":
|
| 1436 |
-
"learning_rate":
|
| 1437 |
-
"loss": 0.
|
| 1438 |
"step": 1750
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
"epoch": 5.48,
|
| 1442 |
-
"grad_norm":
|
| 1443 |
-
"learning_rate":
|
| 1444 |
-
"loss": 0.
|
| 1445 |
"step": 1760
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
"epoch": 5.51,
|
| 1449 |
-
"grad_norm":
|
| 1450 |
-
"learning_rate":
|
| 1451 |
-
"loss": 0.
|
| 1452 |
"step": 1770
|
| 1453 |
},
|
| 1454 |
{
|
| 1455 |
"epoch": 5.55,
|
| 1456 |
-
"grad_norm":
|
| 1457 |
-
"learning_rate":
|
| 1458 |
-
"loss": 0.
|
| 1459 |
"step": 1780
|
| 1460 |
},
|
| 1461 |
{
|
| 1462 |
"epoch": 5.58,
|
| 1463 |
-
"grad_norm":
|
| 1464 |
-
"learning_rate":
|
| 1465 |
-
"loss": 0.
|
| 1466 |
"step": 1790
|
| 1467 |
},
|
| 1468 |
{
|
| 1469 |
"epoch": 5.61,
|
| 1470 |
-
"grad_norm":
|
| 1471 |
-
"learning_rate":
|
| 1472 |
-
"loss": 0.
|
| 1473 |
"step": 1800
|
| 1474 |
},
|
| 1475 |
{
|
| 1476 |
"epoch": 5.61,
|
| 1477 |
-
"eval_accuracy": 0.
|
| 1478 |
-
"eval_f1": 0.
|
| 1479 |
-
"eval_loss": 0.
|
| 1480 |
-
"eval_precision": 0.
|
| 1481 |
-
"eval_recall": 0.
|
| 1482 |
-
"eval_runtime":
|
| 1483 |
-
"eval_samples_per_second":
|
| 1484 |
-
"eval_steps_per_second": 9.
|
| 1485 |
"step": 1800
|
| 1486 |
},
|
| 1487 |
{
|
| 1488 |
"epoch": 5.61,
|
| 1489 |
"step": 1800,
|
| 1490 |
"total_flos": 2.2287694956200755e+18,
|
| 1491 |
-
"train_loss": 0.
|
| 1492 |
-
"train_runtime":
|
| 1493 |
-
"train_samples_per_second":
|
| 1494 |
-
"train_steps_per_second":
|
| 1495 |
}
|
| 1496 |
],
|
| 1497 |
"logging_steps": 10,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.4453237056732178,
|
| 3 |
"best_model_checkpoint": "./vit-lr-cosine-restarts/checkpoint-800",
|
| 4 |
"epoch": 5.607476635514018,
|
| 5 |
"eval_steps": 100,
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.03,
|
| 13 |
+
"grad_norm": 17.940969467163086,
|
| 14 |
+
"learning_rate": 1.125e-05,
|
| 15 |
+
"loss": 2.0172,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.06,
|
| 20 |
+
"grad_norm": 4.308961391448975,
|
| 21 |
+
"learning_rate": 2.375e-05,
|
| 22 |
+
"loss": 1.1159,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.09,
|
| 27 |
+
"grad_norm": 5.38205099105835,
|
| 28 |
+
"learning_rate": 3.625e-05,
|
| 29 |
+
"loss": 1.1398,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.12,
|
| 34 |
+
"grad_norm": 5.569328308105469,
|
| 35 |
+
"learning_rate": 4.875e-05,
|
| 36 |
+
"loss": 1.0508,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.16,
|
| 41 |
+
"grad_norm": 5.870121002197266,
|
| 42 |
+
"learning_rate": 6.125000000000001e-05,
|
| 43 |
+
"loss": 0.8095,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.19,
|
| 48 |
+
"grad_norm": 6.100069046020508,
|
| 49 |
+
"learning_rate": 7.375e-05,
|
| 50 |
+
"loss": 0.8756,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.22,
|
| 55 |
+
"grad_norm": 4.655179023742676,
|
| 56 |
+
"learning_rate": 8.625000000000001e-05,
|
| 57 |
+
"loss": 0.9221,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.25,
|
| 62 |
+
"grad_norm": 4.762995719909668,
|
| 63 |
+
"learning_rate": 9.875000000000002e-05,
|
| 64 |
+
"loss": 0.6852,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.28,
|
| 69 |
+
"grad_norm": 5.309569358825684,
|
| 70 |
+
"learning_rate": 9.99995126719372e-05,
|
| 71 |
+
"loss": 0.6662,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.31,
|
| 76 |
+
"grad_norm": 4.177795886993408,
|
| 77 |
+
"learning_rate": 9.99978280932988e-05,
|
| 78 |
+
"loss": 0.686,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.31,
|
| 83 |
+
"eval_accuracy": 0.7517337031900139,
|
| 84 |
+
"eval_f1": 0.7444977301566197,
|
| 85 |
+
"eval_loss": 0.6707192063331604,
|
| 86 |
+
"eval_precision": 0.7624063091694634,
|
| 87 |
+
"eval_recall": 0.7517337031900139,
|
| 88 |
+
"eval_runtime": 36.8103,
|
| 89 |
+
"eval_samples_per_second": 78.348,
|
| 90 |
+
"eval_steps_per_second": 9.807,
|
| 91 |
"step": 100
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 0.34,
|
| 95 |
+
"grad_norm": 3.6934421062469482,
|
| 96 |
+
"learning_rate": 9.999494028821966e-05,
|
| 97 |
+
"loss": 0.6901,
|
| 98 |
"step": 110
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 0.37,
|
| 102 |
+
"grad_norm": 8.207216262817383,
|
| 103 |
+
"learning_rate": 9.999084932619647e-05,
|
| 104 |
+
"loss": 0.6638,
|
| 105 |
"step": 120
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 0.4,
|
| 109 |
+
"grad_norm": 8.520118713378906,
|
| 110 |
+
"learning_rate": 9.998555530568059e-05,
|
| 111 |
+
"loss": 0.6285,
|
| 112 |
"step": 130
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"epoch": 0.44,
|
| 116 |
+
"grad_norm": 3.053748846054077,
|
| 117 |
+
"learning_rate": 9.997905835407567e-05,
|
| 118 |
+
"loss": 0.5225,
|
| 119 |
"step": 140
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 0.47,
|
| 123 |
+
"grad_norm": 4.431256294250488,
|
| 124 |
+
"learning_rate": 9.997135862773453e-05,
|
| 125 |
+
"loss": 0.6307,
|
| 126 |
"step": 150
|
| 127 |
},
|
| 128 |
{
|
| 129 |
"epoch": 0.5,
|
| 130 |
+
"grad_norm": 7.533073902130127,
|
| 131 |
+
"learning_rate": 9.996245631195555e-05,
|
| 132 |
+
"loss": 0.5959,
|
| 133 |
"step": 160
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 0.53,
|
| 137 |
+
"grad_norm": 5.364689350128174,
|
| 138 |
+
"learning_rate": 9.99523516209781e-05,
|
| 139 |
+
"loss": 0.6344,
|
| 140 |
"step": 170
|
| 141 |
},
|
| 142 |
{
|
| 143 |
"epoch": 0.56,
|
| 144 |
+
"grad_norm": 6.664125919342041,
|
| 145 |
+
"learning_rate": 9.994104479797728e-05,
|
| 146 |
+
"loss": 0.6724,
|
| 147 |
"step": 180
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 0.59,
|
| 151 |
+
"grad_norm": 4.114094257354736,
|
| 152 |
+
"learning_rate": 9.992853611505836e-05,
|
| 153 |
+
"loss": 0.6075,
|
| 154 |
"step": 190
|
| 155 |
},
|
| 156 |
{
|
| 157 |
"epoch": 0.62,
|
| 158 |
+
"grad_norm": 5.480118274688721,
|
| 159 |
+
"learning_rate": 9.991482587324993e-05,
|
| 160 |
+
"loss": 0.4852,
|
| 161 |
"step": 200
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 0.62,
|
| 165 |
+
"eval_accuracy": 0.7704576976421637,
|
| 166 |
+
"eval_f1": 0.7230801818181609,
|
| 167 |
+
"eval_loss": 0.7021857500076294,
|
| 168 |
+
"eval_precision": 0.7857677405887458,
|
| 169 |
+
"eval_recall": 0.7704576976421637,
|
| 170 |
+
"eval_runtime": 37.0615,
|
| 171 |
+
"eval_samples_per_second": 77.817,
|
| 172 |
+
"eval_steps_per_second": 9.741,
|
| 173 |
"step": 200
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 0.65,
|
| 177 |
+
"grad_norm": 4.710792064666748,
|
| 178 |
+
"learning_rate": 9.989991440249686e-05,
|
| 179 |
+
"loss": 0.6202,
|
| 180 |
"step": 210
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 0.69,
|
| 184 |
+
"grad_norm": 8.316452980041504,
|
| 185 |
+
"learning_rate": 9.988380206165225e-05,
|
| 186 |
+
"loss": 0.6816,
|
| 187 |
"step": 220
|
| 188 |
},
|
| 189 |
{
|
| 190 |
"epoch": 0.72,
|
| 191 |
+
"grad_norm": 3.131255626678467,
|
| 192 |
+
"learning_rate": 9.986648923846882e-05,
|
| 193 |
+
"loss": 0.4721,
|
| 194 |
"step": 230
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 0.75,
|
| 198 |
+
"grad_norm": 1.5796229839324951,
|
| 199 |
+
"learning_rate": 9.98479763495896e-05,
|
| 200 |
+
"loss": 0.5474,
|
| 201 |
"step": 240
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 0.78,
|
| 205 |
+
"grad_norm": 5.570051193237305,
|
| 206 |
+
"learning_rate": 9.98282638405379e-05,
|
| 207 |
+
"loss": 0.6763,
|
| 208 |
"step": 250
|
| 209 |
},
|
| 210 |
{
|
| 211 |
"epoch": 0.81,
|
| 212 |
+
"grad_norm": 4.160530090332031,
|
| 213 |
+
"learning_rate": 9.980735218570657e-05,
|
| 214 |
+
"loss": 0.5471,
|
| 215 |
"step": 260
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 0.84,
|
| 219 |
+
"grad_norm": 4.585910320281982,
|
| 220 |
+
"learning_rate": 9.978524188834659e-05,
|
| 221 |
+
"loss": 0.4856,
|
| 222 |
"step": 270
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 0.87,
|
| 226 |
+
"grad_norm": 2.5068447589874268,
|
| 227 |
+
"learning_rate": 9.976193348055496e-05,
|
| 228 |
+
"loss": 0.4724,
|
| 229 |
"step": 280
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"epoch": 0.9,
|
| 233 |
+
"grad_norm": 7.588688373565674,
|
| 234 |
+
"learning_rate": 9.973742752326188e-05,
|
| 235 |
+
"loss": 0.7415,
|
| 236 |
"step": 290
|
| 237 |
},
|
| 238 |
{
|
| 239 |
"epoch": 0.93,
|
| 240 |
+
"grad_norm": 6.468356609344482,
|
| 241 |
+
"learning_rate": 9.971172460621732e-05,
|
| 242 |
+
"loss": 0.7098,
|
| 243 |
"step": 300
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 0.93,
|
| 247 |
+
"eval_accuracy": 0.7995839112343966,
|
| 248 |
+
"eval_f1": 0.7972999747438626,
|
| 249 |
+
"eval_loss": 0.563714325428009,
|
| 250 |
+
"eval_precision": 0.8181443293620774,
|
| 251 |
+
"eval_recall": 0.7995839112343966,
|
| 252 |
+
"eval_runtime": 37.0609,
|
| 253 |
+
"eval_samples_per_second": 77.818,
|
| 254 |
+
"eval_steps_per_second": 9.741,
|
| 255 |
"step": 300
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.97,
|
| 259 |
+
"grad_norm": 4.219545841217041,
|
| 260 |
+
"learning_rate": 9.968482534797669e-05,
|
| 261 |
+
"loss": 0.5762,
|
| 262 |
"step": 310
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 1.0,
|
| 266 |
+
"grad_norm": 5.357555866241455,
|
| 267 |
+
"learning_rate": 9.965673039588614e-05,
|
| 268 |
+
"loss": 0.5181,
|
| 269 |
"step": 320
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 1.03,
|
| 273 |
+
"grad_norm": 4.9291582107543945,
|
| 274 |
+
"learning_rate": 9.962744042606678e-05,
|
| 275 |
+
"loss": 0.422,
|
| 276 |
"step": 330
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 1.06,
|
| 280 |
+
"grad_norm": 5.231845855712891,
|
| 281 |
+
"learning_rate": 9.959695614339857e-05,
|
| 282 |
+
"loss": 0.3889,
|
| 283 |
"step": 340
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 1.09,
|
| 287 |
+
"grad_norm": 3.579317808151245,
|
| 288 |
+
"learning_rate": 9.956527828150326e-05,
|
| 289 |
+
"loss": 0.3912,
|
| 290 |
"step": 350
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 1.12,
|
| 294 |
+
"grad_norm": 3.0033373832702637,
|
| 295 |
+
"learning_rate": 9.95324076027268e-05,
|
| 296 |
+
"loss": 0.3439,
|
| 297 |
"step": 360
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 1.15,
|
| 301 |
+
"grad_norm": 2.9946236610412598,
|
| 302 |
+
"learning_rate": 9.949834489812094e-05,
|
| 303 |
+
"loss": 0.4737,
|
| 304 |
"step": 370
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 1.18,
|
| 308 |
+
"grad_norm": 7.183070182800293,
|
| 309 |
+
"learning_rate": 9.946309098742424e-05,
|
| 310 |
+
"loss": 0.4325,
|
| 311 |
"step": 380
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 1.21,
|
| 315 |
+
"grad_norm": 4.066940784454346,
|
| 316 |
+
"learning_rate": 9.942664671904227e-05,
|
| 317 |
+
"loss": 0.527,
|
| 318 |
"step": 390
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 1.25,
|
| 322 |
+
"grad_norm": 4.427377700805664,
|
| 323 |
+
"learning_rate": 9.938901297002732e-05,
|
| 324 |
+
"loss": 0.4226,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 1.25,
|
| 329 |
+
"eval_accuracy": 0.7621359223300971,
|
| 330 |
+
"eval_f1": 0.7735141412800338,
|
| 331 |
+
"eval_loss": 0.6494001150131226,
|
| 332 |
+
"eval_precision": 0.8136939310866815,
|
| 333 |
+
"eval_recall": 0.7621359223300971,
|
| 334 |
+
"eval_runtime": 37.3436,
|
| 335 |
+
"eval_samples_per_second": 77.229,
|
| 336 |
+
"eval_steps_per_second": 9.667,
|
| 337 |
"step": 400
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 1.28,
|
| 341 |
+
"grad_norm": 3.8335494995117188,
|
| 342 |
+
"learning_rate": 9.935019064605713e-05,
|
| 343 |
+
"loss": 0.4727,
|
| 344 |
"step": 410
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 1.31,
|
| 348 |
+
"grad_norm": 3.420801877975464,
|
| 349 |
+
"learning_rate": 9.931018068141324e-05,
|
| 350 |
+
"loss": 0.4028,
|
| 351 |
"step": 420
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 1.34,
|
| 355 |
+
"grad_norm": 3.57491397857666,
|
| 356 |
+
"learning_rate": 9.926898403895842e-05,
|
| 357 |
+
"loss": 0.5144,
|
| 358 |
"step": 430
|
| 359 |
},
|
| 360 |
{
|
| 361 |
"epoch": 1.37,
|
| 362 |
+
"grad_norm": 2.4824776649475098,
|
| 363 |
+
"learning_rate": 9.92266017101135e-05,
|
| 364 |
+
"loss": 0.3917,
|
| 365 |
"step": 440
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 1.4,
|
| 369 |
+
"grad_norm": 3.9230153560638428,
|
| 370 |
+
"learning_rate": 9.918303471483359e-05,
|
| 371 |
+
"loss": 0.4286,
|
| 372 |
"step": 450
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 1.43,
|
| 376 |
+
"grad_norm": 3.334120512008667,
|
| 377 |
+
"learning_rate": 9.913828410158342e-05,
|
| 378 |
+
"loss": 0.4159,
|
| 379 |
"step": 460
|
| 380 |
},
|
| 381 |
{
|
| 382 |
"epoch": 1.46,
|
| 383 |
+
"grad_norm": 3.502173662185669,
|
| 384 |
+
"learning_rate": 9.909235094731222e-05,
|
| 385 |
+
"loss": 0.3367,
|
| 386 |
"step": 470
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 1.5,
|
| 390 |
+
"grad_norm": 6.164717674255371,
|
| 391 |
+
"learning_rate": 9.90452363574277e-05,
|
| 392 |
+
"loss": 0.3398,
|
| 393 |
"step": 480
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 1.53,
|
| 397 |
+
"grad_norm": 7.483583450317383,
|
| 398 |
+
"learning_rate": 9.899694146576952e-05,
|
| 399 |
+
"loss": 0.3941,
|
| 400 |
"step": 490
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 1.56,
|
| 404 |
+
"grad_norm": 2.83978009223938,
|
| 405 |
+
"learning_rate": 9.8947467434582e-05,
|
| 406 |
+
"loss": 0.3599,
|
| 407 |
"step": 500
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 1.56,
|
| 411 |
+
"eval_accuracy": 0.823509015256588,
|
| 412 |
+
"eval_f1": 0.8108853024782933,
|
| 413 |
+
"eval_loss": 0.5213786363601685,
|
| 414 |
+
"eval_precision": 0.8206648771819358,
|
| 415 |
+
"eval_recall": 0.823509015256588,
|
| 416 |
+
"eval_runtime": 37.1057,
|
| 417 |
+
"eval_samples_per_second": 77.724,
|
| 418 |
+
"eval_steps_per_second": 9.729,
|
| 419 |
"step": 500
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 1.59,
|
| 423 |
+
"grad_norm": 2.34218168258667,
|
| 424 |
+
"learning_rate": 9.889681545448608e-05,
|
| 425 |
+
"loss": 0.4614,
|
| 426 |
"step": 510
|
| 427 |
},
|
| 428 |
{
|
| 429 |
"epoch": 1.62,
|
| 430 |
+
"grad_norm": 4.667867660522461,
|
| 431 |
+
"learning_rate": 9.884498674445075e-05,
|
| 432 |
+
"loss": 0.4268,
|
| 433 |
"step": 520
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 1.65,
|
| 437 |
+
"grad_norm": 8.212017059326172,
|
| 438 |
+
"learning_rate": 9.87919825517637e-05,
|
| 439 |
+
"loss": 0.3964,
|
| 440 |
"step": 530
|
| 441 |
},
|
| 442 |
{
|
| 443 |
"epoch": 1.68,
|
| 444 |
+
"grad_norm": 2.64796781539917,
|
| 445 |
+
"learning_rate": 9.873780415200123e-05,
|
| 446 |
+
"loss": 0.3339,
|
| 447 |
"step": 540
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 1.71,
|
| 451 |
+
"grad_norm": 4.198647975921631,
|
| 452 |
+
"learning_rate": 9.868245284899764e-05,
|
| 453 |
+
"loss": 0.4548,
|
| 454 |
"step": 550
|
| 455 |
},
|
| 456 |
{
|
| 457 |
"epoch": 1.74,
|
| 458 |
+
"grad_norm": 4.222693920135498,
|
| 459 |
+
"learning_rate": 9.862592997481383e-05,
|
| 460 |
+
"loss": 0.3731,
|
| 461 |
"step": 560
|
| 462 |
},
|
| 463 |
{
|
| 464 |
"epoch": 1.78,
|
| 465 |
+
"grad_norm": 3.101301670074463,
|
| 466 |
+
"learning_rate": 9.856823688970525e-05,
|
| 467 |
+
"loss": 0.3133,
|
| 468 |
"step": 570
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 1.81,
|
| 472 |
+
"grad_norm": 3.361762523651123,
|
| 473 |
+
"learning_rate": 9.850937498208906e-05,
|
| 474 |
+
"loss": 0.3255,
|
| 475 |
"step": 580
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 1.84,
|
| 479 |
+
"grad_norm": 5.818238258361816,
|
| 480 |
+
"learning_rate": 9.84493456685109e-05,
|
| 481 |
+
"loss": 0.3941,
|
| 482 |
"step": 590
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 1.87,
|
| 486 |
+
"grad_norm": 2.1035287380218506,
|
| 487 |
+
"learning_rate": 9.838815039361066e-05,
|
| 488 |
+
"loss": 0.3533,
|
| 489 |
"step": 600
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 1.87,
|
| 493 |
+
"eval_accuracy": 0.8273231622746186,
|
| 494 |
+
"eval_f1": 0.8192634434149068,
|
| 495 |
+
"eval_loss": 0.534748911857605,
|
| 496 |
+
"eval_precision": 0.839212218456244,
|
| 497 |
+
"eval_recall": 0.8273231622746186,
|
| 498 |
+
"eval_runtime": 36.5566,
|
| 499 |
+
"eval_samples_per_second": 78.891,
|
| 500 |
+
"eval_steps_per_second": 9.875,
|
| 501 |
"step": 600
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 1.9,
|
| 505 |
+
"grad_norm": 5.71665620803833,
|
| 506 |
+
"learning_rate": 9.832579063008777e-05,
|
| 507 |
+
"loss": 0.4063,
|
| 508 |
"step": 610
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 1.93,
|
| 512 |
+
"grad_norm": 3.79347562789917,
|
| 513 |
+
"learning_rate": 9.826226787866574e-05,
|
| 514 |
+
"loss": 0.427,
|
| 515 |
"step": 620
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 1.96,
|
| 519 |
+
"grad_norm": 2.647468090057373,
|
| 520 |
+
"learning_rate": 9.819758366805607e-05,
|
| 521 |
+
"loss": 0.2353,
|
| 522 |
"step": 630
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 1.99,
|
| 526 |
+
"grad_norm": 3.3511950969696045,
|
| 527 |
+
"learning_rate": 9.813173955492141e-05,
|
| 528 |
+
"loss": 0.3889,
|
| 529 |
"step": 640
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 2.02,
|
| 533 |
+
"grad_norm": 4.286926746368408,
|
| 534 |
+
"learning_rate": 9.806473712383817e-05,
|
| 535 |
+
"loss": 0.2418,
|
| 536 |
"step": 650
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 2.06,
|
| 540 |
+
"grad_norm": 6.692331790924072,
|
| 541 |
+
"learning_rate": 9.79965779872583e-05,
|
| 542 |
+
"loss": 0.2296,
|
| 543 |
"step": 660
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 2.09,
|
| 547 |
+
"grad_norm": 3.424487590789795,
|
| 548 |
+
"learning_rate": 9.792726378547058e-05,
|
| 549 |
+
"loss": 0.1989,
|
| 550 |
"step": 670
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 2.12,
|
| 554 |
+
"grad_norm": 6.994248390197754,
|
| 555 |
+
"learning_rate": 9.785679618656106e-05,
|
| 556 |
+
"loss": 0.2232,
|
| 557 |
"step": 680
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 2.15,
|
| 561 |
+
"grad_norm": 5.3724365234375,
|
| 562 |
+
"learning_rate": 9.778517688637298e-05,
|
| 563 |
+
"loss": 0.2343,
|
| 564 |
"step": 690
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 2.18,
|
| 568 |
+
"grad_norm": 2.975245714187622,
|
| 569 |
+
"learning_rate": 9.77124076084659e-05,
|
| 570 |
+
"loss": 0.1178,
|
| 571 |
"step": 700
|
| 572 |
},
|
| 573 |
{
|
| 574 |
"epoch": 2.18,
|
| 575 |
+
"eval_accuracy": 0.8283633841886269,
|
| 576 |
+
"eval_f1": 0.8277021804405638,
|
| 577 |
+
"eval_loss": 0.5425286889076233,
|
| 578 |
+
"eval_precision": 0.838105973857857,
|
| 579 |
+
"eval_recall": 0.8283633841886269,
|
| 580 |
+
"eval_runtime": 36.4063,
|
| 581 |
+
"eval_samples_per_second": 79.217,
|
| 582 |
+
"eval_steps_per_second": 9.916,
|
| 583 |
"step": 700
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 2.21,
|
| 587 |
+
"grad_norm": 0.8108430504798889,
|
| 588 |
+
"learning_rate": 9.763849010407431e-05,
|
| 589 |
+
"loss": 0.2396,
|
| 590 |
"step": 710
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 2.24,
|
| 594 |
+
"grad_norm": 1.5715973377227783,
|
| 595 |
+
"learning_rate": 9.756342615206538e-05,
|
| 596 |
+
"loss": 0.3591,
|
| 597 |
"step": 720
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 2.27,
|
| 601 |
+
"grad_norm": 5.7021894454956055,
|
| 602 |
+
"learning_rate": 9.748721755889619e-05,
|
| 603 |
+
"loss": 0.3064,
|
| 604 |
"step": 730
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 2.31,
|
| 608 |
+
"grad_norm": 1.4883191585540771,
|
| 609 |
+
"learning_rate": 9.740986615857031e-05,
|
| 610 |
+
"loss": 0.1716,
|
| 611 |
"step": 740
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 2.34,
|
| 615 |
+
"grad_norm": 5.413182735443115,
|
| 616 |
+
"learning_rate": 9.733137381259363e-05,
|
| 617 |
+
"loss": 0.2045,
|
| 618 |
"step": 750
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 2.37,
|
| 622 |
+
"grad_norm": 2.8399062156677246,
|
| 623 |
+
"learning_rate": 9.725174240992947e-05,
|
| 624 |
+
"loss": 0.1557,
|
| 625 |
"step": 760
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 2.4,
|
| 629 |
+
"grad_norm": 8.557807922363281,
|
| 630 |
+
"learning_rate": 9.717097386695331e-05,
|
| 631 |
+
"loss": 0.3044,
|
| 632 |
"step": 770
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 2.43,
|
| 636 |
+
"grad_norm": 4.569642066955566,
|
| 637 |
+
"learning_rate": 9.708907012740649e-05,
|
| 638 |
+
"loss": 0.338,
|
| 639 |
"step": 780
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 2.46,
|
| 643 |
+
"grad_norm": 3.043851852416992,
|
| 644 |
+
"learning_rate": 9.700603316234952e-05,
|
| 645 |
+
"loss": 0.4123,
|
| 646 |
"step": 790
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 2.49,
|
| 650 |
+
"grad_norm": 2.985642194747925,
|
| 651 |
+
"learning_rate": 9.692186497011465e-05,
|
| 652 |
+
"loss": 0.2719,
|
| 653 |
"step": 800
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 2.49,
|
| 657 |
+
"eval_accuracy": 0.8463938973647711,
|
| 658 |
+
"eval_f1": 0.8437993661883203,
|
| 659 |
+
"eval_loss": 0.4453237056732178,
|
| 660 |
+
"eval_precision": 0.8463641738950213,
|
| 661 |
+
"eval_recall": 0.8463938973647711,
|
| 662 |
+
"eval_runtime": 36.5652,
|
| 663 |
+
"eval_samples_per_second": 78.873,
|
| 664 |
+
"eval_steps_per_second": 9.873,
|
| 665 |
"step": 800
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 2.52,
|
| 669 |
+
"grad_norm": 4.473091125488281,
|
| 670 |
+
"learning_rate": 9.683656757625777e-05,
|
| 671 |
+
"loss": 0.2869,
|
| 672 |
"step": 810
|
| 673 |
},
|
| 674 |
{
|
| 675 |
"epoch": 2.55,
|
| 676 |
+
"grad_norm": 7.418603420257568,
|
| 677 |
+
"learning_rate": 9.67501430335096e-05,
|
| 678 |
+
"loss": 0.202,
|
| 679 |
"step": 820
|
| 680 |
},
|
| 681 |
{
|
| 682 |
"epoch": 2.59,
|
| 683 |
+
"grad_norm": 3.375176191329956,
|
| 684 |
+
"learning_rate": 9.666259342172643e-05,
|
| 685 |
+
"loss": 0.1779,
|
| 686 |
"step": 830
|
| 687 |
},
|
| 688 |
{
|
| 689 |
"epoch": 2.62,
|
| 690 |
+
"grad_norm": 4.781165599822998,
|
| 691 |
+
"learning_rate": 9.65739208478399e-05,
|
| 692 |
+
"loss": 0.2148,
|
| 693 |
"step": 840
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 2.65,
|
| 697 |
+
"grad_norm": 0.5722386837005615,
|
| 698 |
+
"learning_rate": 9.648412744580644e-05,
|
| 699 |
+
"loss": 0.1715,
|
| 700 |
"step": 850
|
| 701 |
},
|
| 702 |
{
|
| 703 |
"epoch": 2.68,
|
| 704 |
+
"grad_norm": 5.580469608306885,
|
| 705 |
+
"learning_rate": 9.63932153765558e-05,
|
| 706 |
+
"loss": 0.2103,
|
| 707 |
"step": 860
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 2.71,
|
| 711 |
+
"grad_norm": 8.275124549865723,
|
| 712 |
+
"learning_rate": 9.630118682793917e-05,
|
| 713 |
+
"loss": 0.3448,
|
| 714 |
"step": 870
|
| 715 |
},
|
| 716 |
{
|
| 717 |
"epoch": 2.74,
|
| 718 |
+
"grad_norm": 5.884052753448486,
|
| 719 |
+
"learning_rate": 9.620804401467638e-05,
|
| 720 |
+
"loss": 0.2852,
|
| 721 |
"step": 880
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 2.77,
|
| 725 |
+
"grad_norm": 2.674102783203125,
|
| 726 |
+
"learning_rate": 9.611378917830271e-05,
|
| 727 |
+
"loss": 0.1887,
|
| 728 |
"step": 890
|
| 729 |
},
|
| 730 |
{
|
| 731 |
"epoch": 2.8,
|
| 732 |
+
"grad_norm": 2.95959210395813,
|
| 733 |
+
"learning_rate": 9.601842458711493e-05,
|
| 734 |
+
"loss": 0.1559,
|
| 735 |
"step": 900
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 2.8,
|
| 739 |
+
"eval_accuracy": 0.8325242718446602,
|
| 740 |
+
"eval_f1": 0.8284199470401122,
|
| 741 |
+
"eval_loss": 0.6127275824546814,
|
| 742 |
+
"eval_precision": 0.8566914070001943,
|
| 743 |
+
"eval_recall": 0.8325242718446602,
|
| 744 |
+
"eval_runtime": 37.1542,
|
| 745 |
+
"eval_samples_per_second": 77.623,
|
| 746 |
+
"eval_steps_per_second": 9.716,
|
| 747 |
"step": 900
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 2.83,
|
| 751 |
+
"grad_norm": 3.5971999168395996,
|
| 752 |
+
"learning_rate": 9.592195253611667e-05,
|
| 753 |
+
"loss": 0.3572,
|
| 754 |
"step": 910
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 2.87,
|
| 758 |
+
"grad_norm": 3.3870370388031006,
|
| 759 |
+
"learning_rate": 9.582437534696324e-05,
|
| 760 |
+
"loss": 0.3793,
|
| 761 |
"step": 920
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 2.9,
|
| 765 |
+
"grad_norm": 0.9074994325637817,
|
| 766 |
+
"learning_rate": 9.572569536790572e-05,
|
| 767 |
+
"loss": 0.3371,
|
| 768 |
"step": 930
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 2.93,
|
| 772 |
+
"grad_norm": 6.770932197570801,
|
| 773 |
+
"learning_rate": 9.562591497373448e-05,
|
| 774 |
+
"loss": 0.2833,
|
| 775 |
"step": 940
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 2.96,
|
| 779 |
+
"grad_norm": 6.183658123016357,
|
| 780 |
+
"learning_rate": 9.552503656572196e-05,
|
| 781 |
+
"loss": 0.2882,
|
| 782 |
"step": 950
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 2.99,
|
| 786 |
+
"grad_norm": 5.340375900268555,
|
| 787 |
+
"learning_rate": 9.542306257156502e-05,
|
| 788 |
+
"loss": 0.1809,
|
| 789 |
"step": 960
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 3.02,
|
| 793 |
+
"grad_norm": 4.281813621520996,
|
| 794 |
+
"learning_rate": 9.531999544532633e-05,
|
| 795 |
+
"loss": 0.1301,
|
| 796 |
"step": 970
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 3.05,
|
| 800 |
+
"grad_norm": 2.9234039783477783,
|
| 801 |
+
"learning_rate": 9.521583766737552e-05,
|
| 802 |
+
"loss": 0.088,
|
| 803 |
"step": 980
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 3.08,
|
| 807 |
+
"grad_norm": 0.24170830845832825,
|
| 808 |
+
"learning_rate": 9.511059174432925e-05,
|
| 809 |
+
"loss": 0.0491,
|
| 810 |
"step": 990
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 3.12,
|
| 814 |
+
"grad_norm": 4.773263931274414,
|
| 815 |
+
"learning_rate": 9.500426020899115e-05,
|
| 816 |
+
"loss": 0.1328,
|
| 817 |
"step": 1000
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 3.12,
|
| 821 |
+
"eval_accuracy": 0.8509015256588072,
|
| 822 |
+
"eval_f1": 0.8451195646353651,
|
| 823 |
+
"eval_loss": 0.5303316712379456,
|
| 824 |
+
"eval_precision": 0.845642270599866,
|
| 825 |
+
"eval_recall": 0.8509015256588072,
|
| 826 |
+
"eval_runtime": 36.7597,
|
| 827 |
+
"eval_samples_per_second": 78.456,
|
| 828 |
+
"eval_steps_per_second": 9.821,
|
| 829 |
"step": 1000
|
| 830 |
},
|
| 831 |
{
|
| 832 |
"epoch": 3.15,
|
| 833 |
+
"grad_norm": 1.725915789604187,
|
| 834 |
+
"learning_rate": 9.489684562029066e-05,
|
| 835 |
+
"loss": 0.1083,
|
| 836 |
"step": 1010
|
| 837 |
},
|
| 838 |
{
|
| 839 |
"epoch": 3.18,
|
| 840 |
+
"grad_norm": 4.2252888679504395,
|
| 841 |
+
"learning_rate": 9.47883505632215e-05,
|
| 842 |
+
"loss": 0.1296,
|
| 843 |
"step": 1020
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 3.21,
|
| 847 |
+
"grad_norm": 4.19112491607666,
|
| 848 |
+
"learning_rate": 9.467877764877955e-05,
|
| 849 |
+
"loss": 0.0713,
|
| 850 |
"step": 1030
|
| 851 |
},
|
| 852 |
{
|
| 853 |
"epoch": 3.24,
|
| 854 |
+
"grad_norm": 0.8787875175476074,
|
| 855 |
+
"learning_rate": 9.45681295138999e-05,
|
| 856 |
+
"loss": 0.0602,
|
| 857 |
"step": 1040
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 3.27,
|
| 861 |
+
"grad_norm": 2.9338300228118896,
|
| 862 |
+
"learning_rate": 9.445640882139342e-05,
|
| 863 |
+
"loss": 0.1112,
|
| 864 |
"step": 1050
|
| 865 |
},
|
| 866 |
{
|
| 867 |
"epoch": 3.3,
|
| 868 |
+
"grad_norm": 0.03492557257413864,
|
| 869 |
+
"learning_rate": 9.434361825988276e-05,
|
| 870 |
+
"loss": 0.0632,
|
| 871 |
"step": 1060
|
| 872 |
},
|
| 873 |
{
|
| 874 |
"epoch": 3.33,
|
| 875 |
+
"grad_norm": 7.183565616607666,
|
| 876 |
+
"learning_rate": 9.422976054373753e-05,
|
| 877 |
+
"loss": 0.1271,
|
| 878 |
"step": 1070
|
| 879 |
},
|
| 880 |
{
|
| 881 |
"epoch": 3.36,
|
| 882 |
+
"grad_norm": 5.910800457000732,
|
| 883 |
+
"learning_rate": 9.411483841300905e-05,
|
| 884 |
+
"loss": 0.1384,
|
| 885 |
"step": 1080
|
| 886 |
},
|
| 887 |
{
|
| 888 |
"epoch": 3.4,
|
| 889 |
+
"grad_norm": 4.911332607269287,
|
| 890 |
+
"learning_rate": 9.399885463336437e-05,
|
| 891 |
+
"loss": 0.0607,
|
| 892 |
"step": 1090
|
| 893 |
},
|
| 894 |
{
|
| 895 |
"epoch": 3.43,
|
| 896 |
+
"grad_norm": 1.9047012329101562,
|
| 897 |
+
"learning_rate": 9.388181199601974e-05,
|
| 898 |
+
"loss": 0.1756,
|
| 899 |
"step": 1100
|
| 900 |
},
|
| 901 |
{
|
| 902 |
"epoch": 3.43,
|
| 903 |
+
"eval_accuracy": 0.8321775312066574,
|
| 904 |
+
"eval_f1": 0.8151306059680461,
|
| 905 |
+
"eval_loss": 0.7960126996040344,
|
| 906 |
+
"eval_precision": 0.8366372545968512,
|
| 907 |
+
"eval_recall": 0.8321775312066574,
|
| 908 |
+
"eval_runtime": 36.834,
|
| 909 |
+
"eval_samples_per_second": 78.297,
|
| 910 |
+
"eval_steps_per_second": 9.801,
|
| 911 |
"step": 1100
|
| 912 |
},
|
| 913 |
{
|
| 914 |
"epoch": 3.46,
|
| 915 |
+
"grad_norm": 0.7878803610801697,
|
| 916 |
+
"learning_rate": 9.376371331767345e-05,
|
| 917 |
+
"loss": 0.1006,
|
| 918 |
"step": 1110
|
| 919 |
},
|
| 920 |
{
|
| 921 |
"epoch": 3.49,
|
| 922 |
+
"grad_norm": 1.042022705078125,
|
| 923 |
+
"learning_rate": 9.364456144043798e-05,
|
| 924 |
+
"loss": 0.1516,
|
| 925 |
"step": 1120
|
| 926 |
},
|
| 927 |
{
|
| 928 |
"epoch": 3.52,
|
| 929 |
+
"grad_norm": 1.4984287023544312,
|
| 930 |
+
"learning_rate": 9.35243592317717e-05,
|
| 931 |
+
"loss": 0.0771,
|
| 932 |
"step": 1130
|
| 933 |
},
|
| 934 |
{
|
| 935 |
"epoch": 3.55,
|
| 936 |
+
"grad_norm": 7.682912349700928,
|
| 937 |
+
"learning_rate": 9.340310958440976e-05,
|
| 938 |
+
"loss": 0.0898,
|
| 939 |
"step": 1140
|
| 940 |
},
|
| 941 |
{
|
| 942 |
"epoch": 3.58,
|
| 943 |
+
"grad_norm": 4.866548538208008,
|
| 944 |
+
"learning_rate": 9.328081541629453e-05,
|
| 945 |
+
"loss": 0.1182,
|
| 946 |
"step": 1150
|
| 947 |
},
|
| 948 |
{
|
| 949 |
"epoch": 3.61,
|
| 950 |
+
"grad_norm": 2.1378111839294434,
|
| 951 |
+
"learning_rate": 9.315747967050541e-05,
|
| 952 |
+
"loss": 0.2255,
|
| 953 |
"step": 1160
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 3.64,
|
| 957 |
+
"grad_norm": 1.4697102308273315,
|
| 958 |
+
"learning_rate": 9.303310531518793e-05,
|
| 959 |
+
"loss": 0.1011,
|
| 960 |
"step": 1170
|
| 961 |
},
|
| 962 |
{
|
| 963 |
"epoch": 3.68,
|
| 964 |
+
"grad_norm": 5.105794429779053,
|
| 965 |
+
"learning_rate": 9.290769534348236e-05,
|
| 966 |
+
"loss": 0.1298,
|
| 967 |
"step": 1180
|
| 968 |
},
|
| 969 |
{
|
| 970 |
"epoch": 3.71,
|
| 971 |
+
"grad_norm": 5.116852760314941,
|
| 972 |
+
"learning_rate": 9.278125277345168e-05,
|
| 973 |
+
"loss": 0.1145,
|
| 974 |
"step": 1190
|
| 975 |
},
|
| 976 |
{
|
| 977 |
"epoch": 3.74,
|
| 978 |
+
"grad_norm": 1.9126471281051636,
|
| 979 |
+
"learning_rate": 9.265378064800895e-05,
|
| 980 |
+
"loss": 0.3582,
|
| 981 |
"step": 1200
|
| 982 |
},
|
| 983 |
{
|
| 984 |
"epoch": 3.74,
|
| 985 |
+
"eval_accuracy": 0.834257975034674,
|
| 986 |
+
"eval_f1": 0.824916890066515,
|
| 987 |
+
"eval_loss": 0.6675512790679932,
|
| 988 |
+
"eval_precision": 0.8284494824114729,
|
| 989 |
+
"eval_recall": 0.834257975034674,
|
| 990 |
+
"eval_runtime": 36.0917,
|
| 991 |
+
"eval_samples_per_second": 79.908,
|
| 992 |
+
"eval_steps_per_second": 10.002,
|
| 993 |
"step": 1200
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 3.77,
|
| 997 |
+
"grad_norm": 4.828185081481934,
|
| 998 |
+
"learning_rate": 9.252528203484403e-05,
|
| 999 |
+
"loss": 0.1843,
|
| 1000 |
"step": 1210
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
"epoch": 3.8,
|
| 1004 |
+
"grad_norm": 2.7517149448394775,
|
| 1005 |
+
"learning_rate": 9.239576002634984e-05,
|
| 1006 |
+
"loss": 0.1066,
|
| 1007 |
"step": 1220
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
"epoch": 3.83,
|
| 1011 |
+
"grad_norm": 3.601691246032715,
|
| 1012 |
+
"learning_rate": 9.226521773954791e-05,
|
| 1013 |
+
"loss": 0.1121,
|
| 1014 |
"step": 1230
|
| 1015 |
},
|
| 1016 |
{
|
| 1017 |
"epoch": 3.86,
|
| 1018 |
+
"grad_norm": 0.0293317511677742,
|
| 1019 |
+
"learning_rate": 9.21336583160133e-05,
|
| 1020 |
+
"loss": 0.1822,
|
| 1021 |
"step": 1240
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 3.89,
|
| 1025 |
+
"grad_norm": 0.6248491406440735,
|
| 1026 |
+
"learning_rate": 9.200108492179906e-05,
|
| 1027 |
+
"loss": 0.1261,
|
| 1028 |
"step": 1250
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
"epoch": 3.93,
|
| 1032 |
+
"grad_norm": 0.1484900414943695,
|
| 1033 |
+
"learning_rate": 9.186750074736009e-05,
|
| 1034 |
+
"loss": 0.1224,
|
| 1035 |
"step": 1260
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
"epoch": 3.96,
|
| 1039 |
+
"grad_norm": 2.4208881855010986,
|
| 1040 |
+
"learning_rate": 9.17329090074762e-05,
|
| 1041 |
+
"loss": 0.1018,
|
| 1042 |
"step": 1270
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
"epoch": 3.99,
|
| 1046 |
+
"grad_norm": 0.208229199051857,
|
| 1047 |
+
"learning_rate": 9.159731294117492e-05,
|
| 1048 |
+
"loss": 0.1453,
|
| 1049 |
"step": 1280
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
"epoch": 4.02,
|
| 1053 |
+
"grad_norm": 0.03745197877287865,
|
| 1054 |
+
"learning_rate": 9.146071581165345e-05,
|
| 1055 |
+
"loss": 0.1056,
|
| 1056 |
"step": 1290
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
"epoch": 4.05,
|
| 1060 |
+
"grad_norm": 1.308124303817749,
|
| 1061 |
+
"learning_rate": 9.132312090620011e-05,
|
| 1062 |
+
"loss": 0.025,
|
| 1063 |
"step": 1300
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
"epoch": 4.05,
|
| 1067 |
+
"eval_accuracy": 0.8474341192787794,
|
| 1068 |
+
"eval_f1": 0.8476771584783079,
|
| 1069 |
+
"eval_loss": 0.5981015563011169,
|
| 1070 |
+
"eval_precision": 0.859867898706205,
|
| 1071 |
+
"eval_recall": 0.8474341192787794,
|
| 1072 |
+
"eval_runtime": 36.0165,
|
| 1073 |
+
"eval_samples_per_second": 80.074,
|
| 1074 |
+
"eval_steps_per_second": 10.023,
|
| 1075 |
"step": 1300
|
| 1076 |
},
|
| 1077 |
{
|
| 1078 |
"epoch": 4.08,
|
| 1079 |
+
"grad_norm": 0.0758899599313736,
|
| 1080 |
+
"learning_rate": 9.118453153611532e-05,
|
| 1081 |
+
"loss": 0.0298,
|
| 1082 |
"step": 1310
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
"epoch": 4.11,
|
| 1086 |
+
"grad_norm": 0.28617605566978455,
|
| 1087 |
+
"learning_rate": 9.104495103663187e-05,
|
| 1088 |
+
"loss": 0.017,
|
| 1089 |
"step": 1320
|
| 1090 |
},
|
| 1091 |
{
|
| 1092 |
"epoch": 4.14,
|
| 1093 |
+
"grad_norm": 0.19124433398246765,
|
| 1094 |
+
"learning_rate": 9.090438276683457e-05,
|
| 1095 |
+
"loss": 0.017,
|
| 1096 |
"step": 1330
|
| 1097 |
},
|
| 1098 |
{
|
| 1099 |
"epoch": 4.17,
|
| 1100 |
+
"grad_norm": 0.31906023621559143,
|
| 1101 |
+
"learning_rate": 9.07628301095796e-05,
|
| 1102 |
+
"loss": 0.0196,
|
| 1103 |
"step": 1340
|
| 1104 |
},
|
| 1105 |
{
|
| 1106 |
"epoch": 4.21,
|
| 1107 |
+
"grad_norm": 7.870569229125977,
|
| 1108 |
+
"learning_rate": 9.062029647141289e-05,
|
| 1109 |
+
"loss": 0.0685,
|
| 1110 |
"step": 1350
|
| 1111 |
},
|
| 1112 |
{
|
| 1113 |
"epoch": 4.24,
|
| 1114 |
+
"grad_norm": 5.755252361297607,
|
| 1115 |
+
"learning_rate": 9.04767852824883e-05,
|
| 1116 |
+
"loss": 0.0266,
|
| 1117 |
"step": 1360
|
| 1118 |
},
|
| 1119 |
{
|
| 1120 |
"epoch": 4.27,
|
| 1121 |
+
"grad_norm": 0.05481214076280594,
|
| 1122 |
+
"learning_rate": 9.0332299996485e-05,
|
| 1123 |
+
"loss": 0.0594,
|
| 1124 |
"step": 1370
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
"epoch": 4.3,
|
| 1128 |
+
"grad_norm": 5.236385345458984,
|
| 1129 |
+
"learning_rate": 9.018684409052436e-05,
|
| 1130 |
+
"loss": 0.0999,
|
| 1131 |
"step": 1380
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
"epoch": 4.33,
|
| 1135 |
+
"grad_norm": 5.066316604614258,
|
| 1136 |
+
"learning_rate": 9.004042106508625e-05,
|
| 1137 |
+
"loss": 0.0612,
|
| 1138 |
"step": 1390
|
| 1139 |
},
|
| 1140 |
{
|
| 1141 |
"epoch": 4.36,
|
| 1142 |
+
"grad_norm": 0.9376081824302673,
|
| 1143 |
+
"learning_rate": 8.989303444392487e-05,
|
| 1144 |
+
"loss": 0.042,
|
| 1145 |
"step": 1400
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
"epoch": 4.36,
|
| 1149 |
+
"eval_accuracy": 0.8162274618585298,
|
| 1150 |
+
"eval_f1": 0.8241381969601037,
|
| 1151 |
+
"eval_loss": 0.8095719814300537,
|
| 1152 |
+
"eval_precision": 0.8477203881282387,
|
| 1153 |
+
"eval_recall": 0.8162274618585298,
|
| 1154 |
+
"eval_runtime": 36.3685,
|
| 1155 |
+
"eval_samples_per_second": 79.299,
|
| 1156 |
+
"eval_steps_per_second": 9.926,
|
| 1157 |
"step": 1400
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 4.39,
|
| 1161 |
+
"grad_norm": 0.013413701206445694,
|
| 1162 |
+
"learning_rate": 8.974468777398388e-05,
|
| 1163 |
+
"loss": 0.1021,
|
| 1164 |
"step": 1410
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
"epoch": 4.42,
|
| 1168 |
+
"grad_norm": 7.129204750061035,
|
| 1169 |
+
"learning_rate": 8.959538462531108e-05,
|
| 1170 |
+
"loss": 0.1356,
|
| 1171 |
"step": 1420
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 4.45,
|
| 1175 |
+
"grad_norm": 3.306025505065918,
|
| 1176 |
+
"learning_rate": 8.944512859097245e-05,
|
| 1177 |
+
"loss": 0.0191,
|
| 1178 |
"step": 1430
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 4.49,
|
| 1182 |
+
"grad_norm": 0.022593187168240547,
|
| 1183 |
+
"learning_rate": 8.929392328696574e-05,
|
| 1184 |
+
"loss": 0.0448,
|
| 1185 |
"step": 1440
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 4.52,
|
| 1189 |
+
"grad_norm": 4.569972991943359,
|
| 1190 |
+
"learning_rate": 8.914177235213341e-05,
|
| 1191 |
+
"loss": 0.073,
|
| 1192 |
"step": 1450
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
"epoch": 4.55,
|
| 1196 |
+
"grad_norm": 4.406450271606445,
|
| 1197 |
+
"learning_rate": 8.898867944807507e-05,
|
| 1198 |
+
"loss": 0.0672,
|
| 1199 |
"step": 1460
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 4.58,
|
| 1203 |
+
"grad_norm": 7.7769904136657715,
|
| 1204 |
+
"learning_rate": 8.883464825905934e-05,
|
| 1205 |
+
"loss": 0.0947,
|
| 1206 |
"step": 1470
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
"epoch": 4.61,
|
| 1210 |
+
"grad_norm": 9.705739974975586,
|
| 1211 |
+
"learning_rate": 8.867968249193526e-05,
|
| 1212 |
+
"loss": 0.0344,
|
| 1213 |
"step": 1480
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 4.64,
|
| 1217 |
+
"grad_norm": 1.0004101991653442,
|
| 1218 |
+
"learning_rate": 8.852378587604297e-05,
|
| 1219 |
+
"loss": 0.0877,
|
| 1220 |
"step": 1490
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
"epoch": 4.67,
|
| 1224 |
+
"grad_norm": 0.8759760856628418,
|
| 1225 |
+
"learning_rate": 8.836696216312405e-05,
|
| 1226 |
+
"loss": 0.05,
|
| 1227 |
"step": 1500
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 4.67,
|
| 1231 |
+
"eval_accuracy": 0.841886269070735,
|
| 1232 |
+
"eval_f1": 0.8341416187793224,
|
| 1233 |
+
"eval_loss": 0.7948272228240967,
|
| 1234 |
+
"eval_precision": 0.8474305891416275,
|
| 1235 |
+
"eval_recall": 0.841886269070735,
|
| 1236 |
+
"eval_runtime": 37.2997,
|
| 1237 |
+
"eval_samples_per_second": 77.32,
|
| 1238 |
+
"eval_steps_per_second": 9.678,
|
| 1239 |
"step": 1500
|
| 1240 |
},
|
| 1241 |
{
|
| 1242 |
"epoch": 4.7,
|
| 1243 |
+
"grad_norm": 0.008059758692979813,
|
| 1244 |
+
"learning_rate": 8.82092151272312e-05,
|
| 1245 |
+
"loss": 0.0939,
|
| 1246 |
"step": 1510
|
| 1247 |
},
|
| 1248 |
{
|
| 1249 |
"epoch": 4.74,
|
| 1250 |
+
"grad_norm": 0.09355029463768005,
|
| 1251 |
+
"learning_rate": 8.80505485646374e-05,
|
| 1252 |
+
"loss": 0.03,
|
| 1253 |
"step": 1520
|
| 1254 |
},
|
| 1255 |
{
|
| 1256 |
"epoch": 4.77,
|
| 1257 |
+
"grad_norm": 0.4395085871219635,
|
| 1258 |
+
"learning_rate": 8.78909662937446e-05,
|
| 1259 |
+
"loss": 0.1189,
|
| 1260 |
"step": 1530
|
| 1261 |
},
|
| 1262 |
{
|
| 1263 |
"epoch": 4.8,
|
| 1264 |
+
"grad_norm": 0.6208884119987488,
|
| 1265 |
+
"learning_rate": 8.773047215499176e-05,
|
| 1266 |
+
"loss": 0.06,
|
| 1267 |
"step": 1540
|
| 1268 |
},
|
| 1269 |
{
|
| 1270 |
"epoch": 4.83,
|
| 1271 |
+
"grad_norm": 1.5753854513168335,
|
| 1272 |
+
"learning_rate": 8.756907001076249e-05,
|
| 1273 |
+
"loss": 0.0517,
|
| 1274 |
"step": 1550
|
| 1275 |
},
|
| 1276 |
{
|
| 1277 |
"epoch": 4.86,
|
| 1278 |
+
"grad_norm": 5.975317001342773,
|
| 1279 |
+
"learning_rate": 8.740676374529206e-05,
|
| 1280 |
+
"loss": 0.1042,
|
| 1281 |
"step": 1560
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
"epoch": 4.89,
|
| 1285 |
+
"grad_norm": 3.8262839317321777,
|
| 1286 |
+
"learning_rate": 8.724355726457395e-05,
|
| 1287 |
+
"loss": 0.0716,
|
| 1288 |
"step": 1570
|
| 1289 |
},
|
| 1290 |
{
|
| 1291 |
"epoch": 4.92,
|
| 1292 |
+
"grad_norm": 2.5249273777008057,
|
| 1293 |
+
"learning_rate": 8.707945449626583e-05,
|
| 1294 |
+
"loss": 0.0254,
|
| 1295 |
"step": 1580
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
"epoch": 4.95,
|
| 1299 |
+
"grad_norm": 10.16901969909668,
|
| 1300 |
+
"learning_rate": 8.691445938959504e-05,
|
| 1301 |
+
"loss": 0.1318,
|
| 1302 |
"step": 1590
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
"epoch": 4.98,
|
| 1306 |
+
"grad_norm": 1.012184977531433,
|
| 1307 |
+
"learning_rate": 8.674857591526355e-05,
|
| 1308 |
+
"loss": 0.028,
|
| 1309 |
"step": 1600
|
| 1310 |
},
|
| 1311 |
{
|
| 1312 |
"epoch": 4.98,
|
| 1313 |
+
"eval_accuracy": 0.8457004160887656,
|
| 1314 |
+
"eval_f1": 0.8476462818721602,
|
| 1315 |
+
"eval_loss": 0.6741925477981567,
|
| 1316 |
+
"eval_precision": 0.8558392490201036,
|
| 1317 |
+
"eval_recall": 0.8457004160887656,
|
| 1318 |
+
"eval_runtime": 37.79,
|
| 1319 |
+
"eval_samples_per_second": 76.316,
|
| 1320 |
+
"eval_steps_per_second": 9.553,
|
| 1321 |
"step": 1600
|
| 1322 |
},
|
| 1323 |
{
|
| 1324 |
"epoch": 5.02,
|
| 1325 |
+
"grad_norm": 1.2411854267120361,
|
| 1326 |
+
"learning_rate": 8.658180806535243e-05,
|
| 1327 |
+
"loss": 0.0079,
|
| 1328 |
"step": 1610
|
| 1329 |
},
|
| 1330 |
{
|
| 1331 |
"epoch": 5.05,
|
| 1332 |
+
"grad_norm": 8.111499786376953,
|
| 1333 |
+
"learning_rate": 8.641415985322571e-05,
|
| 1334 |
+
"loss": 0.0455,
|
| 1335 |
"step": 1620
|
| 1336 |
},
|
| 1337 |
{
|
| 1338 |
"epoch": 5.08,
|
| 1339 |
+
"grad_norm": 0.05484266206622124,
|
| 1340 |
+
"learning_rate": 8.624563531343393e-05,
|
| 1341 |
+
"loss": 0.0255,
|
| 1342 |
"step": 1630
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
"epoch": 5.11,
|
| 1346 |
+
"grad_norm": 0.6857353448867798,
|
| 1347 |
+
"learning_rate": 8.607623850161686e-05,
|
| 1348 |
+
"loss": 0.0115,
|
| 1349 |
"step": 1640
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
"epoch": 5.14,
|
| 1353 |
+
"grad_norm": 0.38402843475341797,
|
| 1354 |
+
"learning_rate": 8.590597349440604e-05,
|
| 1355 |
+
"loss": 0.0171,
|
| 1356 |
"step": 1650
|
| 1357 |
},
|
| 1358 |
{
|
| 1359 |
"epoch": 5.17,
|
| 1360 |
+
"grad_norm": 0.07074743509292603,
|
| 1361 |
+
"learning_rate": 8.573484438932666e-05,
|
| 1362 |
+
"loss": 0.0067,
|
| 1363 |
"step": 1660
|
| 1364 |
},
|
| 1365 |
{
|
| 1366 |
"epoch": 5.2,
|
| 1367 |
+
"grad_norm": 0.1260824352502823,
|
| 1368 |
+
"learning_rate": 8.556285530469887e-05,
|
| 1369 |
+
"loss": 0.009,
|
| 1370 |
"step": 1670
|
| 1371 |
},
|
| 1372 |
{
|
| 1373 |
"epoch": 5.23,
|
| 1374 |
+
"grad_norm": 0.23438212275505066,
|
| 1375 |
+
"learning_rate": 8.539001037953876e-05,
|
| 1376 |
+
"loss": 0.0232,
|
| 1377 |
"step": 1680
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
"epoch": 5.26,
|
| 1381 |
+
"grad_norm": 0.19910460710525513,
|
| 1382 |
+
"learning_rate": 8.521631377345869e-05,
|
| 1383 |
+
"loss": 0.0022,
|
| 1384 |
"step": 1690
|
| 1385 |
},
|
| 1386 |
{
|
| 1387 |
"epoch": 5.3,
|
| 1388 |
+
"grad_norm": 0.016467662528157234,
|
| 1389 |
+
"learning_rate": 8.50417696665672e-05,
|
| 1390 |
+
"loss": 0.0048,
|
| 1391 |
"step": 1700
|
| 1392 |
},
|
| 1393 |
{
|
| 1394 |
"epoch": 5.3,
|
| 1395 |
+
"eval_accuracy": 0.8484743411927878,
|
| 1396 |
+
"eval_f1": 0.8499629260872099,
|
| 1397 |
+
"eval_loss": 0.7832539081573486,
|
| 1398 |
+
"eval_precision": 0.8576584679191768,
|
| 1399 |
+
"eval_recall": 0.8484743411927878,
|
| 1400 |
+
"eval_runtime": 37.0076,
|
| 1401 |
+
"eval_samples_per_second": 77.93,
|
| 1402 |
+
"eval_steps_per_second": 9.755,
|
| 1403 |
"step": 1700
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
"epoch": 5.33,
|
| 1407 |
+
"grad_norm": 0.006973025389015675,
|
| 1408 |
+
"learning_rate": 8.486638225936848e-05,
|
| 1409 |
+
"loss": 0.073,
|
| 1410 |
"step": 1710
|
| 1411 |
},
|
| 1412 |
{
|
| 1413 |
"epoch": 5.36,
|
| 1414 |
+
"grad_norm": 0.29455187916755676,
|
| 1415 |
+
"learning_rate": 8.469015577266115e-05,
|
| 1416 |
+
"loss": 0.002,
|
| 1417 |
"step": 1720
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
"epoch": 5.39,
|
| 1421 |
+
"grad_norm": 0.06881581246852875,
|
| 1422 |
+
"learning_rate": 8.451309444743682e-05,
|
| 1423 |
+
"loss": 0.0479,
|
| 1424 |
"step": 1730
|
| 1425 |
},
|
| 1426 |
{
|
| 1427 |
"epoch": 5.42,
|
| 1428 |
+
"grad_norm": 0.03852876275777817,
|
| 1429 |
+
"learning_rate": 8.433520254477793e-05,
|
| 1430 |
+
"loss": 0.0821,
|
| 1431 |
"step": 1740
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
"epoch": 5.45,
|
| 1435 |
+
"grad_norm": 0.0018428952898830175,
|
| 1436 |
+
"learning_rate": 8.415648434575529e-05,
|
| 1437 |
+
"loss": 0.0375,
|
| 1438 |
"step": 1750
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
"epoch": 5.48,
|
| 1442 |
+
"grad_norm": 0.003056368324905634,
|
| 1443 |
+
"learning_rate": 8.397694415132495e-05,
|
| 1444 |
+
"loss": 0.0884,
|
| 1445 |
"step": 1760
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
"epoch": 5.51,
|
| 1449 |
+
"grad_norm": 1.8021758794784546,
|
| 1450 |
+
"learning_rate": 8.379658628222478e-05,
|
| 1451 |
+
"loss": 0.0091,
|
| 1452 |
"step": 1770
|
| 1453 |
},
|
| 1454 |
{
|
| 1455 |
"epoch": 5.55,
|
| 1456 |
+
"grad_norm": 0.1937793642282486,
|
| 1457 |
+
"learning_rate": 8.361541507887045e-05,
|
| 1458 |
+
"loss": 0.0101,
|
| 1459 |
"step": 1780
|
| 1460 |
},
|
| 1461 |
{
|
| 1462 |
"epoch": 5.58,
|
| 1463 |
+
"grad_norm": 2.138684034347534,
|
| 1464 |
+
"learning_rate": 8.343343490125102e-05,
|
| 1465 |
+
"loss": 0.1135,
|
| 1466 |
"step": 1790
|
| 1467 |
},
|
| 1468 |
{
|
| 1469 |
"epoch": 5.61,
|
| 1470 |
+
"grad_norm": 0.781872034072876,
|
| 1471 |
+
"learning_rate": 8.325065012882392e-05,
|
| 1472 |
+
"loss": 0.0324,
|
| 1473 |
"step": 1800
|
| 1474 |
},
|
| 1475 |
{
|
| 1476 |
"epoch": 5.61,
|
| 1477 |
+
"eval_accuracy": 0.8533287101248266,
|
| 1478 |
+
"eval_f1": 0.8511397162148655,
|
| 1479 |
+
"eval_loss": 0.7405093312263489,
|
| 1480 |
+
"eval_precision": 0.8523403828700276,
|
| 1481 |
+
"eval_recall": 0.8533287101248266,
|
| 1482 |
+
"eval_runtime": 37.0368,
|
| 1483 |
+
"eval_samples_per_second": 77.869,
|
| 1484 |
+
"eval_steps_per_second": 9.747,
|
| 1485 |
"step": 1800
|
| 1486 |
},
|
| 1487 |
{
|
| 1488 |
"epoch": 5.61,
|
| 1489 |
"step": 1800,
|
| 1490 |
"total_flos": 2.2287694956200755e+18,
|
| 1491 |
+
"train_loss": 0.2811500767639114,
|
| 1492 |
+
"train_runtime": 1301.746,
|
| 1493 |
+
"train_samples_per_second": 393.932,
|
| 1494 |
+
"train_steps_per_second": 24.659
|
| 1495 |
}
|
| 1496 |
],
|
| 1497 |
"logging_steps": 10,
|