🍻 cheers
Browse files- README.md +4 -3
- all_results.json +11 -6
- eval_results.json +8 -0
- runs/Apr26_01-09-42_fe7d46d3b18e/events.out.tfevents.1714095382.fe7d46d3b18e.21275.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +1332 -312
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# finetuned-electrical-images
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 16 |
|
| 17 |
# finetuned-electrical-images
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the Electrical_components(VIT) dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.3634
|
| 22 |
+
- Accuracy: 0.8977
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
CHANGED
|
@@ -1,8 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8976897689768977,
|
| 4 |
+
"eval_loss": 0.36343997716903687,
|
| 5 |
+
"eval_runtime": 8.0693,
|
| 6 |
+
"eval_samples_per_second": 75.1,
|
| 7 |
+
"eval_steps_per_second": 9.418,
|
| 8 |
+
"total_flos": 2.659620429501235e+18,
|
| 9 |
+
"train_loss": 0.3051174351919529,
|
| 10 |
+
"train_runtime": 1549.0078,
|
| 11 |
+
"train_samples_per_second": 22.156,
|
| 12 |
+
"train_steps_per_second": 1.388
|
| 13 |
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8976897689768977,
|
| 4 |
+
"eval_loss": 0.36343997716903687,
|
| 5 |
+
"eval_runtime": 8.0693,
|
| 6 |
+
"eval_samples_per_second": 75.1,
|
| 7 |
+
"eval_steps_per_second": 9.418
|
| 8 |
+
}
|
runs/Apr26_01-09-42_fe7d46d3b18e/events.out.tfevents.1714095382.fe7d46d3b18e.21275.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1683b9d5a5405b35cef77ed6620d3c2484766ab33174b0a1b09ad3b7d08094c3
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 2.659620429501235e+18,
|
| 4 |
+
"train_loss": 0.3051174351919529,
|
| 5 |
+
"train_runtime": 1549.0078,
|
| 6 |
+
"train_samples_per_second": 22.156,
|
| 7 |
+
"train_steps_per_second": 1.388
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,703 +1,1723 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "finetuned-electrical-images/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.046511627906976744,
|
| 13 |
-
"grad_norm": 2.
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09302325581395349,
|
| 20 |
-
"grad_norm": 2.
|
| 21 |
-
"learning_rate": 0.
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.13953488372093023,
|
| 27 |
-
"grad_norm": 2.
|
| 28 |
-
"learning_rate": 0.
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.18604651162790697,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss":
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.23255813953488372,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.27906976744186046,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.32558139534883723,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 0.
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.37209302325581395,
|
| 62 |
-
"grad_norm": 2.
|
| 63 |
-
"learning_rate": 0.
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.4186046511627907,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.46511627906976744,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.46511627906976744,
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_loss": 0.
|
| 85 |
-
"eval_runtime":
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second": 9.
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.5116279069767442,
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate": 0.
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.5581395348837209,
|
| 99 |
-
"grad_norm": 5.
|
| 100 |
-
"learning_rate": 0.
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 0.6046511627906976,
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate": 0.
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 0.6511627906976745,
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate": 0.
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.6976744186046512,
|
| 120 |
-
"grad_norm": 3.
|
| 121 |
-
"learning_rate": 0.
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.7441860465116279,
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate": 0.
|
| 129 |
-
"loss": 0.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.7906976744186046,
|
| 134 |
-
"grad_norm": 2.
|
| 135 |
-
"learning_rate": 0.
|
| 136 |
-
"loss": 0.
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.8372093023255814,
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate": 0.
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 0.8837209302325582,
|
| 148 |
-
"grad_norm": 3.
|
| 149 |
-
"learning_rate": 0.
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 0.9302325581395349,
|
| 155 |
-
"grad_norm": 1.
|
| 156 |
-
"learning_rate": 0.
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 0.9302325581395349,
|
| 162 |
-
"eval_accuracy": 0.
|
| 163 |
-
"eval_loss": 0.
|
| 164 |
-
"eval_runtime":
|
| 165 |
-
"eval_samples_per_second":
|
| 166 |
-
"eval_steps_per_second":
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.9767441860465116,
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate": 0.
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 1.0232558139534884,
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate": 0.
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 1.069767441860465,
|
| 185 |
-
"grad_norm":
|
| 186 |
-
"learning_rate": 0.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 1.1162790697674418,
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate": 0.
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 1.1627906976744187,
|
| 199 |
-
"grad_norm":
|
| 200 |
-
"learning_rate": 0.
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 1.2093023255813953,
|
| 206 |
-
"grad_norm":
|
| 207 |
-
"learning_rate": 0.
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 1.255813953488372,
|
| 213 |
-
"grad_norm":
|
| 214 |
-
"learning_rate": 0.
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 1.302325581395349,
|
| 220 |
-
"grad_norm": 1.
|
| 221 |
-
"learning_rate": 0.
|
| 222 |
-
"loss": 0.
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 1.3488372093023255,
|
| 227 |
-
"grad_norm":
|
| 228 |
-
"learning_rate": 0.
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 1.3953488372093024,
|
| 234 |
-
"grad_norm":
|
| 235 |
-
"learning_rate": 0.
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 1.3953488372093024,
|
| 241 |
-
"eval_accuracy": 0.
|
| 242 |
-
"eval_loss": 0.
|
| 243 |
-
"eval_runtime": 8.
|
| 244 |
-
"eval_samples_per_second": 73.
|
| 245 |
-
"eval_steps_per_second": 9.
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 1.441860465116279,
|
| 250 |
-
"grad_norm":
|
| 251 |
-
"learning_rate": 0.
|
| 252 |
-
"loss": 0.
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 1.4883720930232558,
|
| 257 |
-
"grad_norm":
|
| 258 |
-
"learning_rate": 0.
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 1.5348837209302326,
|
| 264 |
-
"grad_norm":
|
| 265 |
-
"learning_rate": 0.
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 1.5813953488372094,
|
| 271 |
-
"grad_norm":
|
| 272 |
-
"learning_rate": 0.
|
| 273 |
-
"loss": 0.
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 1.627906976744186,
|
| 278 |
-
"grad_norm":
|
| 279 |
-
"learning_rate": 0.
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 1.6744186046511627,
|
| 285 |
-
"grad_norm":
|
| 286 |
-
"learning_rate": 0.
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 1.7209302325581395,
|
| 292 |
-
"grad_norm": 2.
|
| 293 |
-
"learning_rate": 0.
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 1.7674418604651163,
|
| 299 |
-
"grad_norm":
|
| 300 |
-
"learning_rate": 0.
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 1.8139534883720931,
|
| 306 |
-
"grad_norm":
|
| 307 |
-
"learning_rate": 0.
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 1.8604651162790697,
|
| 313 |
-
"grad_norm": 2.
|
| 314 |
-
"learning_rate": 0.
|
| 315 |
-
"loss": 0.
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 1.8604651162790697,
|
| 320 |
-
"eval_accuracy": 0.
|
| 321 |
-
"eval_loss": 0.
|
| 322 |
-
"eval_runtime": 8.
|
| 323 |
-
"eval_samples_per_second": 74.
|
| 324 |
-
"eval_steps_per_second": 9.
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 1.9069767441860463,
|
| 329 |
-
"grad_norm":
|
| 330 |
-
"learning_rate": 0.
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 1.9534883720930232,
|
| 336 |
-
"grad_norm": 3.
|
| 337 |
-
"learning_rate": 0.
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 2.0,
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate": 0.
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 2.046511627906977,
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 2.0930232558139537,
|
| 357 |
-
"grad_norm": 2.
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 2.13953488372093,
|
| 364 |
-
"grad_norm": 2.
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 2.186046511627907,
|
| 371 |
-
"grad_norm":
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 2.2325581395348837,
|
| 378 |
-
"grad_norm":
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 2.2790697674418605,
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 2.3255813953488373,
|
| 392 |
-
"grad_norm":
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 2.3255813953488373,
|
| 399 |
-
"eval_accuracy": 0.
|
| 400 |
-
"eval_loss": 0.
|
| 401 |
-
"eval_runtime": 8.
|
| 402 |
-
"eval_samples_per_second":
|
| 403 |
-
"eval_steps_per_second": 9.
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 2.3720930232558137,
|
| 408 |
-
"grad_norm": 4.
|
| 409 |
-
"learning_rate":
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 2.4186046511627906,
|
| 415 |
-
"grad_norm":
|
| 416 |
-
"learning_rate":
|
| 417 |
-
"loss": 0.
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 2.4651162790697674,
|
| 422 |
-
"grad_norm": 3.
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 2.511627906976744,
|
| 429 |
-
"grad_norm":
|
| 430 |
-
"learning_rate":
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 2.558139534883721,
|
| 436 |
-
"grad_norm":
|
| 437 |
-
"learning_rate":
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 2.604651162790698,
|
| 443 |
-
"grad_norm": 1.
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 2.6511627906976747,
|
| 450 |
-
"grad_norm":
|
| 451 |
-
"learning_rate":
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 2.697674418604651,
|
| 457 |
-
"grad_norm":
|
| 458 |
-
"learning_rate":
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 2.744186046511628,
|
| 464 |
-
"grad_norm":
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 2.7906976744186047,
|
| 471 |
-
"grad_norm":
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 2.7906976744186047,
|
| 478 |
-
"eval_accuracy": 0.
|
| 479 |
-
"eval_loss": 0.
|
| 480 |
-
"eval_runtime":
|
| 481 |
-
"eval_samples_per_second":
|
| 482 |
-
"eval_steps_per_second": 9.
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 2.8372093023255816,
|
| 487 |
-
"grad_norm":
|
| 488 |
-
"learning_rate":
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 2.883720930232558,
|
| 494 |
-
"grad_norm":
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 2.9302325581395348,
|
| 501 |
-
"grad_norm":
|
| 502 |
-
"learning_rate":
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 2.9767441860465116,
|
| 508 |
-
"grad_norm":
|
| 509 |
-
"learning_rate":
|
| 510 |
-
"loss": 0.
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 3.0232558139534884,
|
| 515 |
-
"grad_norm":
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 3.0697674418604652,
|
| 522 |
-
"grad_norm":
|
| 523 |
-
"learning_rate":
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 3.116279069767442,
|
| 529 |
-
"grad_norm":
|
| 530 |
-
"learning_rate":
|
| 531 |
-
"loss": 0.
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 3.1627906976744184,
|
| 536 |
-
"grad_norm":
|
| 537 |
-
"learning_rate":
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 3.2093023255813953,
|
| 543 |
-
"grad_norm":
|
| 544 |
-
"learning_rate":
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 3.255813953488372,
|
| 550 |
-
"grad_norm":
|
| 551 |
-
"learning_rate":
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 3.255813953488372,
|
| 557 |
-
"eval_accuracy": 0.
|
| 558 |
-
"eval_loss": 0.
|
| 559 |
-
"eval_runtime": 8.
|
| 560 |
-
"eval_samples_per_second": 75.
|
| 561 |
-
"eval_steps_per_second": 9.
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 3.302325581395349,
|
| 566 |
-
"grad_norm":
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 3.3488372093023258,
|
| 573 |
-
"grad_norm":
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 3.395348837209302,
|
| 580 |
-
"grad_norm": 3.
|
| 581 |
-
"learning_rate":
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 3.441860465116279,
|
| 587 |
-
"grad_norm":
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 3.488372093023256,
|
| 594 |
-
"grad_norm":
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 3.5348837209302326,
|
| 601 |
-
"grad_norm": 3.
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 3.5813953488372094,
|
| 608 |
-
"grad_norm":
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 3.6279069767441863,
|
| 615 |
-
"grad_norm":
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 3.6744186046511627,
|
| 622 |
-
"grad_norm": 1.
|
| 623 |
-
"learning_rate":
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 3.7209302325581395,
|
| 629 |
-
"grad_norm": 5.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 3.7209302325581395,
|
| 636 |
-
"eval_accuracy": 0.
|
| 637 |
-
"eval_loss": 0.
|
| 638 |
-
"eval_runtime": 8.
|
| 639 |
-
"eval_samples_per_second":
|
| 640 |
-
"eval_steps_per_second":
|
| 641 |
"step": 800
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 3.7674418604651163,
|
| 645 |
-
"grad_norm":
|
| 646 |
-
"learning_rate":
|
| 647 |
-
"loss": 0.
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
"epoch": 3.813953488372093,
|
| 652 |
-
"grad_norm": 2.
|
| 653 |
-
"learning_rate":
|
| 654 |
-
"loss": 0.
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 3.8604651162790695,
|
| 659 |
-
"grad_norm": 2.
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 3.9069767441860463,
|
| 666 |
-
"grad_norm":
|
| 667 |
-
"learning_rate":
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 3.953488372093023,
|
| 673 |
-
"grad_norm":
|
| 674 |
-
"learning_rate":
|
| 675 |
-
"loss": 0.
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 4.0,
|
| 680 |
-
"grad_norm":
|
| 681 |
-
"learning_rate": 0.
|
| 682 |
-
"loss": 0.
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
-
"epoch": 4.
|
| 687 |
-
"
|
| 688 |
-
"
|
| 689 |
-
"
|
| 690 |
-
"
|
| 691 |
-
|
| 692 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 693 |
}
|
| 694 |
],
|
| 695 |
"logging_steps": 10,
|
| 696 |
-
"max_steps":
|
| 697 |
"num_input_tokens_seen": 0,
|
| 698 |
-
"num_train_epochs":
|
| 699 |
"save_steps": 100,
|
| 700 |
-
"total_flos":
|
| 701 |
"train_batch_size": 16,
|
| 702 |
"trial_name": null,
|
| 703 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.36343997716903687,
|
| 3 |
+
"best_model_checkpoint": "finetuned-electrical-images/checkpoint-1300",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 2150,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.046511627906976744,
|
| 13 |
+
"grad_norm": 2.1625218391418457,
|
| 14 |
+
"learning_rate": 0.00019906976744186048,
|
| 15 |
+
"loss": 1.5809,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09302325581395349,
|
| 20 |
+
"grad_norm": 2.723552703857422,
|
| 21 |
+
"learning_rate": 0.00019813953488372096,
|
| 22 |
+
"loss": 1.2765,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.13953488372093023,
|
| 27 |
+
"grad_norm": 2.43442702293396,
|
| 28 |
+
"learning_rate": 0.0001972093023255814,
|
| 29 |
+
"loss": 1.1166,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.18604651162790697,
|
| 34 |
+
"grad_norm": 3.831533670425415,
|
| 35 |
+
"learning_rate": 0.00019627906976744185,
|
| 36 |
+
"loss": 1.0432,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.23255813953488372,
|
| 41 |
+
"grad_norm": 1.6888493299484253,
|
| 42 |
+
"learning_rate": 0.00019534883720930232,
|
| 43 |
+
"loss": 0.8645,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.27906976744186046,
|
| 48 |
+
"grad_norm": 3.6029133796691895,
|
| 49 |
+
"learning_rate": 0.0001944186046511628,
|
| 50 |
+
"loss": 0.8017,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.32558139534883723,
|
| 55 |
+
"grad_norm": 3.071220636367798,
|
| 56 |
+
"learning_rate": 0.00019348837209302326,
|
| 57 |
+
"loss": 0.8024,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.37209302325581395,
|
| 62 |
+
"grad_norm": 2.4523532390594482,
|
| 63 |
+
"learning_rate": 0.00019255813953488374,
|
| 64 |
+
"loss": 0.7961,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.4186046511627907,
|
| 69 |
+
"grad_norm": 2.8159303665161133,
|
| 70 |
+
"learning_rate": 0.0001916279069767442,
|
| 71 |
+
"loss": 0.7968,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.46511627906976744,
|
| 76 |
+
"grad_norm": 2.5400731563568115,
|
| 77 |
+
"learning_rate": 0.00019069767441860466,
|
| 78 |
+
"loss": 0.7236,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.46511627906976744,
|
| 83 |
+
"eval_accuracy": 0.8102310231023102,
|
| 84 |
+
"eval_loss": 0.6395960450172424,
|
| 85 |
+
"eval_runtime": 8.3162,
|
| 86 |
+
"eval_samples_per_second": 72.87,
|
| 87 |
+
"eval_steps_per_second": 9.139,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.5116279069767442,
|
| 92 |
+
"grad_norm": 5.171610355377197,
|
| 93 |
+
"learning_rate": 0.00018976744186046513,
|
| 94 |
+
"loss": 0.6643,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.5581395348837209,
|
| 99 |
+
"grad_norm": 5.763847827911377,
|
| 100 |
+
"learning_rate": 0.00018883720930232557,
|
| 101 |
+
"loss": 0.6579,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 0.6046511627906976,
|
| 106 |
+
"grad_norm": 2.429523229598999,
|
| 107 |
+
"learning_rate": 0.00018790697674418605,
|
| 108 |
+
"loss": 0.5872,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 0.6511627906976745,
|
| 113 |
+
"grad_norm": 2.6132380962371826,
|
| 114 |
+
"learning_rate": 0.00018697674418604652,
|
| 115 |
+
"loss": 0.6796,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.6976744186046512,
|
| 120 |
+
"grad_norm": 3.3510539531707764,
|
| 121 |
+
"learning_rate": 0.000186046511627907,
|
| 122 |
+
"loss": 0.5738,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.7441860465116279,
|
| 127 |
+
"grad_norm": 3.596841812133789,
|
| 128 |
+
"learning_rate": 0.00018511627906976744,
|
| 129 |
+
"loss": 0.5647,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.7906976744186046,
|
| 134 |
+
"grad_norm": 2.6907944679260254,
|
| 135 |
+
"learning_rate": 0.0001841860465116279,
|
| 136 |
+
"loss": 0.5161,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.8372093023255814,
|
| 141 |
+
"grad_norm": 2.8201518058776855,
|
| 142 |
+
"learning_rate": 0.00018325581395348838,
|
| 143 |
+
"loss": 0.5416,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 0.8837209302325582,
|
| 148 |
+
"grad_norm": 3.49650239944458,
|
| 149 |
+
"learning_rate": 0.00018232558139534886,
|
| 150 |
+
"loss": 0.5873,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 0.9302325581395349,
|
| 155 |
+
"grad_norm": 1.33900785446167,
|
| 156 |
+
"learning_rate": 0.0001813953488372093,
|
| 157 |
+
"loss": 0.7243,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 0.9302325581395349,
|
| 162 |
+
"eval_accuracy": 0.8333333333333334,
|
| 163 |
+
"eval_loss": 0.5123782753944397,
|
| 164 |
+
"eval_runtime": 9.3797,
|
| 165 |
+
"eval_samples_per_second": 64.608,
|
| 166 |
+
"eval_steps_per_second": 8.103,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.9767441860465116,
|
| 171 |
+
"grad_norm": 2.2079968452453613,
|
| 172 |
+
"learning_rate": 0.00018046511627906977,
|
| 173 |
+
"loss": 0.5871,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 1.0232558139534884,
|
| 178 |
+
"grad_norm": 2.7433505058288574,
|
| 179 |
+
"learning_rate": 0.00017953488372093025,
|
| 180 |
+
"loss": 0.5605,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 1.069767441860465,
|
| 185 |
+
"grad_norm": 4.832902908325195,
|
| 186 |
+
"learning_rate": 0.0001786046511627907,
|
| 187 |
+
"loss": 0.4689,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 1.1162790697674418,
|
| 192 |
+
"grad_norm": 3.20340633392334,
|
| 193 |
+
"learning_rate": 0.00017767441860465117,
|
| 194 |
+
"loss": 0.4492,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 1.1627906976744187,
|
| 199 |
+
"grad_norm": 3.986785650253296,
|
| 200 |
+
"learning_rate": 0.00017674418604651164,
|
| 201 |
+
"loss": 0.477,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 1.2093023255813953,
|
| 206 |
+
"grad_norm": 3.679204225540161,
|
| 207 |
+
"learning_rate": 0.0001758139534883721,
|
| 208 |
+
"loss": 0.4499,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 1.255813953488372,
|
| 213 |
+
"grad_norm": 5.091402530670166,
|
| 214 |
+
"learning_rate": 0.00017488372093023258,
|
| 215 |
+
"loss": 0.4271,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 1.302325581395349,
|
| 220 |
+
"grad_norm": 1.7408082485198975,
|
| 221 |
+
"learning_rate": 0.00017395348837209303,
|
| 222 |
+
"loss": 0.4805,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 1.3488372093023255,
|
| 227 |
+
"grad_norm": 1.776224136352539,
|
| 228 |
+
"learning_rate": 0.00017302325581395348,
|
| 229 |
+
"loss": 0.5031,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 1.3953488372093024,
|
| 234 |
+
"grad_norm": 1.4141583442687988,
|
| 235 |
+
"learning_rate": 0.00017209302325581395,
|
| 236 |
+
"loss": 0.4288,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 1.3953488372093024,
|
| 241 |
+
"eval_accuracy": 0.863036303630363,
|
| 242 |
+
"eval_loss": 0.45135730504989624,
|
| 243 |
+
"eval_runtime": 8.2539,
|
| 244 |
+
"eval_samples_per_second": 73.42,
|
| 245 |
+
"eval_steps_per_second": 9.208,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 1.441860465116279,
|
| 250 |
+
"grad_norm": 3.1988017559051514,
|
| 251 |
+
"learning_rate": 0.00017116279069767442,
|
| 252 |
+
"loss": 0.4641,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 1.4883720930232558,
|
| 257 |
+
"grad_norm": 3.4049489498138428,
|
| 258 |
+
"learning_rate": 0.0001702325581395349,
|
| 259 |
+
"loss": 0.5539,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 1.5348837209302326,
|
| 264 |
+
"grad_norm": 2.879892349243164,
|
| 265 |
+
"learning_rate": 0.00016930232558139537,
|
| 266 |
+
"loss": 0.4818,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 1.5813953488372094,
|
| 271 |
+
"grad_norm": 3.1155309677124023,
|
| 272 |
+
"learning_rate": 0.00016837209302325584,
|
| 273 |
+
"loss": 0.5773,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 1.627906976744186,
|
| 278 |
+
"grad_norm": 5.9107866287231445,
|
| 279 |
+
"learning_rate": 0.00016744186046511629,
|
| 280 |
+
"loss": 0.5665,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 1.6744186046511627,
|
| 285 |
+
"grad_norm": 3.242166519165039,
|
| 286 |
+
"learning_rate": 0.00016651162790697673,
|
| 287 |
+
"loss": 0.4483,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 1.7209302325581395,
|
| 292 |
+
"grad_norm": 2.6073851585388184,
|
| 293 |
+
"learning_rate": 0.0001655813953488372,
|
| 294 |
+
"loss": 0.3956,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 1.7674418604651163,
|
| 299 |
+
"grad_norm": 2.7050387859344482,
|
| 300 |
+
"learning_rate": 0.00016465116279069768,
|
| 301 |
+
"loss": 0.5734,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 1.8139534883720931,
|
| 306 |
+
"grad_norm": 4.9598259925842285,
|
| 307 |
+
"learning_rate": 0.00016372093023255815,
|
| 308 |
+
"loss": 0.6565,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 1.8604651162790697,
|
| 313 |
+
"grad_norm": 2.172515630722046,
|
| 314 |
+
"learning_rate": 0.00016279069767441862,
|
| 315 |
+
"loss": 0.5744,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 1.8604651162790697,
|
| 320 |
+
"eval_accuracy": 0.8102310231023102,
|
| 321 |
+
"eval_loss": 0.6154166460037231,
|
| 322 |
+
"eval_runtime": 8.1716,
|
| 323 |
+
"eval_samples_per_second": 74.159,
|
| 324 |
+
"eval_steps_per_second": 9.3,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 1.9069767441860463,
|
| 329 |
+
"grad_norm": 1.5915610790252686,
|
| 330 |
+
"learning_rate": 0.00016186046511627907,
|
| 331 |
+
"loss": 0.6044,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 1.9534883720930232,
|
| 336 |
+
"grad_norm": 3.225754976272583,
|
| 337 |
+
"learning_rate": 0.00016093023255813954,
|
| 338 |
+
"loss": 0.5832,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 2.0,
|
| 343 |
+
"grad_norm": 5.616280555725098,
|
| 344 |
+
"learning_rate": 0.00016,
|
| 345 |
+
"loss": 0.3845,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 2.046511627906977,
|
| 350 |
+
"grad_norm": 2.0786819458007812,
|
| 351 |
+
"learning_rate": 0.00015906976744186046,
|
| 352 |
+
"loss": 0.4349,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 2.0930232558139537,
|
| 357 |
+
"grad_norm": 2.8910179138183594,
|
| 358 |
+
"learning_rate": 0.00015813953488372093,
|
| 359 |
+
"loss": 0.4858,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 2.13953488372093,
|
| 364 |
+
"grad_norm": 2.445451259613037,
|
| 365 |
+
"learning_rate": 0.0001572093023255814,
|
| 366 |
+
"loss": 0.3341,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 2.186046511627907,
|
| 371 |
+
"grad_norm": 3.979907751083374,
|
| 372 |
+
"learning_rate": 0.00015627906976744188,
|
| 373 |
+
"loss": 0.4102,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 2.2325581395348837,
|
| 378 |
+
"grad_norm": 3.332590103149414,
|
| 379 |
+
"learning_rate": 0.00015534883720930232,
|
| 380 |
+
"loss": 0.4107,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 2.2790697674418605,
|
| 385 |
+
"grad_norm": 3.54638671875,
|
| 386 |
+
"learning_rate": 0.0001544186046511628,
|
| 387 |
+
"loss": 0.3151,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 2.3255813953488373,
|
| 392 |
+
"grad_norm": 5.199957847595215,
|
| 393 |
+
"learning_rate": 0.00015348837209302327,
|
| 394 |
+
"loss": 0.4077,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 2.3255813953488373,
|
| 399 |
+
"eval_accuracy": 0.8613861386138614,
|
| 400 |
+
"eval_loss": 0.46117761731147766,
|
| 401 |
+
"eval_runtime": 8.1068,
|
| 402 |
+
"eval_samples_per_second": 74.752,
|
| 403 |
+
"eval_steps_per_second": 9.375,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 2.3720930232558137,
|
| 408 |
+
"grad_norm": 4.220390796661377,
|
| 409 |
+
"learning_rate": 0.00015255813953488374,
|
| 410 |
+
"loss": 0.4136,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 2.4186046511627906,
|
| 415 |
+
"grad_norm": 3.7662441730499268,
|
| 416 |
+
"learning_rate": 0.0001516279069767442,
|
| 417 |
+
"loss": 0.4278,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 2.4651162790697674,
|
| 422 |
+
"grad_norm": 3.1768009662628174,
|
| 423 |
+
"learning_rate": 0.00015069767441860466,
|
| 424 |
+
"loss": 0.3468,
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 2.511627906976744,
|
| 429 |
+
"grad_norm": 4.637648105621338,
|
| 430 |
+
"learning_rate": 0.0001497674418604651,
|
| 431 |
+
"loss": 0.4745,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 2.558139534883721,
|
| 436 |
+
"grad_norm": 4.119113445281982,
|
| 437 |
+
"learning_rate": 0.00014883720930232558,
|
| 438 |
+
"loss": 0.3556,
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 2.604651162790698,
|
| 443 |
+
"grad_norm": 1.932075023651123,
|
| 444 |
+
"learning_rate": 0.00014790697674418605,
|
| 445 |
+
"loss": 0.3924,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 2.6511627906976747,
|
| 450 |
+
"grad_norm": 1.4679527282714844,
|
| 451 |
+
"learning_rate": 0.00014697674418604652,
|
| 452 |
+
"loss": 0.492,
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 2.697674418604651,
|
| 457 |
+
"grad_norm": 3.493939161300659,
|
| 458 |
+
"learning_rate": 0.000146046511627907,
|
| 459 |
+
"loss": 0.4785,
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 2.744186046511628,
|
| 464 |
+
"grad_norm": 5.127101898193359,
|
| 465 |
+
"learning_rate": 0.00014511627906976747,
|
| 466 |
+
"loss": 0.3809,
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 2.7906976744186047,
|
| 471 |
+
"grad_norm": 4.6078925132751465,
|
| 472 |
+
"learning_rate": 0.00014418604651162791,
|
| 473 |
+
"loss": 0.496,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 2.7906976744186047,
|
| 478 |
+
"eval_accuracy": 0.8729372937293729,
|
| 479 |
+
"eval_loss": 0.43590766191482544,
|
| 480 |
+
"eval_runtime": 7.9386,
|
| 481 |
+
"eval_samples_per_second": 76.336,
|
| 482 |
+
"eval_steps_per_second": 9.573,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 2.8372093023255816,
|
| 487 |
+
"grad_norm": 2.777296781539917,
|
| 488 |
+
"learning_rate": 0.00014325581395348836,
|
| 489 |
+
"loss": 0.3873,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 2.883720930232558,
|
| 494 |
+
"grad_norm": 3.4130568504333496,
|
| 495 |
+
"learning_rate": 0.00014232558139534883,
|
| 496 |
+
"loss": 0.4155,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 2.9302325581395348,
|
| 501 |
+
"grad_norm": 3.3363895416259766,
|
| 502 |
+
"learning_rate": 0.0001413953488372093,
|
| 503 |
+
"loss": 0.4497,
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 2.9767441860465116,
|
| 508 |
+
"grad_norm": 2.6814212799072266,
|
| 509 |
+
"learning_rate": 0.00014046511627906978,
|
| 510 |
+
"loss": 0.3669,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 3.0232558139534884,
|
| 515 |
+
"grad_norm": 4.29351282119751,
|
| 516 |
+
"learning_rate": 0.00013953488372093025,
|
| 517 |
+
"loss": 0.3421,
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 3.0697674418604652,
|
| 522 |
+
"grad_norm": 2.0479118824005127,
|
| 523 |
+
"learning_rate": 0.00013860465116279072,
|
| 524 |
+
"loss": 0.3297,
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 3.116279069767442,
|
| 529 |
+
"grad_norm": 1.3133976459503174,
|
| 530 |
+
"learning_rate": 0.00013767441860465117,
|
| 531 |
+
"loss": 0.2098,
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 3.1627906976744184,
|
| 536 |
+
"grad_norm": 0.7082284092903137,
|
| 537 |
+
"learning_rate": 0.00013674418604651162,
|
| 538 |
+
"loss": 0.4151,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 3.2093023255813953,
|
| 543 |
+
"grad_norm": 3.3736181259155273,
|
| 544 |
+
"learning_rate": 0.0001358139534883721,
|
| 545 |
+
"loss": 0.2073,
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 3.255813953488372,
|
| 550 |
+
"grad_norm": 7.461119174957275,
|
| 551 |
+
"learning_rate": 0.00013488372093023256,
|
| 552 |
+
"loss": 0.3446,
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 3.255813953488372,
|
| 557 |
+
"eval_accuracy": 0.8696369636963697,
|
| 558 |
+
"eval_loss": 0.42761871218681335,
|
| 559 |
+
"eval_runtime": 8.0701,
|
| 560 |
+
"eval_samples_per_second": 75.092,
|
| 561 |
+
"eval_steps_per_second": 9.417,
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 3.302325581395349,
|
| 566 |
+
"grad_norm": 1.2921802997589111,
|
| 567 |
+
"learning_rate": 0.00013395348837209303,
|
| 568 |
+
"loss": 0.3886,
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 3.3488372093023258,
|
| 573 |
+
"grad_norm": 2.9924850463867188,
|
| 574 |
+
"learning_rate": 0.0001330232558139535,
|
| 575 |
+
"loss": 0.2564,
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 3.395348837209302,
|
| 580 |
+
"grad_norm": 3.2593166828155518,
|
| 581 |
+
"learning_rate": 0.00013209302325581395,
|
| 582 |
+
"loss": 0.404,
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 3.441860465116279,
|
| 587 |
+
"grad_norm": 2.4605319499969482,
|
| 588 |
+
"learning_rate": 0.00013116279069767442,
|
| 589 |
+
"loss": 0.5092,
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 3.488372093023256,
|
| 594 |
+
"grad_norm": 2.1411373615264893,
|
| 595 |
+
"learning_rate": 0.0001302325581395349,
|
| 596 |
+
"loss": 0.3141,
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 3.5348837209302326,
|
| 601 |
+
"grad_norm": 3.7920267581939697,
|
| 602 |
+
"learning_rate": 0.00012930232558139534,
|
| 603 |
+
"loss": 0.2392,
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 3.5813953488372094,
|
| 608 |
+
"grad_norm": 2.0424487590789795,
|
| 609 |
+
"learning_rate": 0.00012837209302325582,
|
| 610 |
+
"loss": 0.3916,
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 3.6279069767441863,
|
| 615 |
+
"grad_norm": 3.782304286956787,
|
| 616 |
+
"learning_rate": 0.0001274418604651163,
|
| 617 |
+
"loss": 0.2629,
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 3.6744186046511627,
|
| 622 |
+
"grad_norm": 1.529003381729126,
|
| 623 |
+
"learning_rate": 0.00012651162790697676,
|
| 624 |
+
"loss": 0.3023,
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 3.7209302325581395,
|
| 629 |
+
"grad_norm": 5.351986885070801,
|
| 630 |
+
"learning_rate": 0.0001255813953488372,
|
| 631 |
+
"loss": 0.3347,
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 3.7209302325581395,
|
| 636 |
+
"eval_accuracy": 0.8795379537953796,
|
| 637 |
+
"eval_loss": 0.4259110689163208,
|
| 638 |
+
"eval_runtime": 8.618,
|
| 639 |
+
"eval_samples_per_second": 70.318,
|
| 640 |
+
"eval_steps_per_second": 8.819,
|
| 641 |
"step": 800
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 3.7674418604651163,
|
| 645 |
+
"grad_norm": 4.942080974578857,
|
| 646 |
+
"learning_rate": 0.00012465116279069768,
|
| 647 |
+
"loss": 0.2278,
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
"epoch": 3.813953488372093,
|
| 652 |
+
"grad_norm": 2.3388428688049316,
|
| 653 |
+
"learning_rate": 0.00012372093023255815,
|
| 654 |
+
"loss": 0.1785,
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 3.8604651162790695,
|
| 659 |
+
"grad_norm": 2.0570669174194336,
|
| 660 |
+
"learning_rate": 0.00012279069767441863,
|
| 661 |
+
"loss": 0.3932,
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 3.9069767441860463,
|
| 666 |
+
"grad_norm": 5.025542736053467,
|
| 667 |
+
"learning_rate": 0.00012186046511627907,
|
| 668 |
+
"loss": 0.3624,
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 3.953488372093023,
|
| 673 |
+
"grad_norm": 2.5320193767547607,
|
| 674 |
+
"learning_rate": 0.00012093023255813953,
|
| 675 |
+
"loss": 0.2873,
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 4.0,
|
| 680 |
+
"grad_norm": 7.442944526672363,
|
| 681 |
+
"learning_rate": 0.00012,
|
| 682 |
+
"loss": 0.3133,
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
+
"epoch": 4.046511627906977,
|
| 687 |
+
"grad_norm": 5.330275535583496,
|
| 688 |
+
"learning_rate": 0.00011906976744186048,
|
| 689 |
+
"loss": 0.2818,
|
| 690 |
+
"step": 870
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 4.093023255813954,
|
| 694 |
+
"grad_norm": 4.320654392242432,
|
| 695 |
+
"learning_rate": 0.00011813953488372094,
|
| 696 |
+
"loss": 0.4261,
|
| 697 |
+
"step": 880
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 4.1395348837209305,
|
| 701 |
+
"grad_norm": 3.309112310409546,
|
| 702 |
+
"learning_rate": 0.00011720930232558141,
|
| 703 |
+
"loss": 0.2859,
|
| 704 |
+
"step": 890
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 4.186046511627907,
|
| 708 |
+
"grad_norm": 2.0252270698547363,
|
| 709 |
+
"learning_rate": 0.00011627906976744187,
|
| 710 |
+
"loss": 0.3868,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 4.186046511627907,
|
| 715 |
+
"eval_accuracy": 0.8547854785478548,
|
| 716 |
+
"eval_loss": 0.46416693925857544,
|
| 717 |
+
"eval_runtime": 8.3175,
|
| 718 |
+
"eval_samples_per_second": 72.859,
|
| 719 |
+
"eval_steps_per_second": 9.137,
|
| 720 |
+
"step": 900
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 4.232558139534884,
|
| 724 |
+
"grad_norm": 0.6425607204437256,
|
| 725 |
+
"learning_rate": 0.00011534883720930234,
|
| 726 |
+
"loss": 0.224,
|
| 727 |
+
"step": 910
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 4.27906976744186,
|
| 731 |
+
"grad_norm": 1.0721712112426758,
|
| 732 |
+
"learning_rate": 0.00011441860465116279,
|
| 733 |
+
"loss": 0.2792,
|
| 734 |
+
"step": 920
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 4.325581395348837,
|
| 738 |
+
"grad_norm": 3.4592926502227783,
|
| 739 |
+
"learning_rate": 0.00011348837209302326,
|
| 740 |
+
"loss": 0.1881,
|
| 741 |
+
"step": 930
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 4.372093023255814,
|
| 745 |
+
"grad_norm": 3.121467351913452,
|
| 746 |
+
"learning_rate": 0.00011255813953488372,
|
| 747 |
+
"loss": 0.2873,
|
| 748 |
+
"step": 940
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 4.4186046511627906,
|
| 752 |
+
"grad_norm": 1.8312969207763672,
|
| 753 |
+
"learning_rate": 0.00011162790697674419,
|
| 754 |
+
"loss": 0.2052,
|
| 755 |
+
"step": 950
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 4.465116279069767,
|
| 759 |
+
"grad_norm": 2.4395313262939453,
|
| 760 |
+
"learning_rate": 0.00011069767441860466,
|
| 761 |
+
"loss": 0.2376,
|
| 762 |
+
"step": 960
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 4.511627906976744,
|
| 766 |
+
"grad_norm": 4.029181480407715,
|
| 767 |
+
"learning_rate": 0.00010976744186046512,
|
| 768 |
+
"loss": 0.2401,
|
| 769 |
+
"step": 970
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 4.558139534883721,
|
| 773 |
+
"grad_norm": 5.82197904586792,
|
| 774 |
+
"learning_rate": 0.0001088372093023256,
|
| 775 |
+
"loss": 0.3262,
|
| 776 |
+
"step": 980
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 4.604651162790698,
|
| 780 |
+
"grad_norm": 3.295616388320923,
|
| 781 |
+
"learning_rate": 0.00010790697674418607,
|
| 782 |
+
"loss": 0.233,
|
| 783 |
+
"step": 990
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 4.651162790697675,
|
| 787 |
+
"grad_norm": 4.5378899574279785,
|
| 788 |
+
"learning_rate": 0.00010697674418604651,
|
| 789 |
+
"loss": 0.36,
|
| 790 |
+
"step": 1000
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 4.651162790697675,
|
| 794 |
+
"eval_accuracy": 0.8696369636963697,
|
| 795 |
+
"eval_loss": 0.4241912066936493,
|
| 796 |
+
"eval_runtime": 11.6827,
|
| 797 |
+
"eval_samples_per_second": 51.872,
|
| 798 |
+
"eval_steps_per_second": 6.505,
|
| 799 |
+
"step": 1000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 4.6976744186046515,
|
| 803 |
+
"grad_norm": 2.949002504348755,
|
| 804 |
+
"learning_rate": 0.00010604651162790697,
|
| 805 |
+
"loss": 0.1716,
|
| 806 |
+
"step": 1010
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 4.7441860465116275,
|
| 810 |
+
"grad_norm": 3.2645139694213867,
|
| 811 |
+
"learning_rate": 0.00010511627906976745,
|
| 812 |
+
"loss": 0.2992,
|
| 813 |
+
"step": 1020
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 4.790697674418604,
|
| 817 |
+
"grad_norm": 0.3114562928676605,
|
| 818 |
+
"learning_rate": 0.0001041860465116279,
|
| 819 |
+
"loss": 0.268,
|
| 820 |
+
"step": 1030
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 4.837209302325581,
|
| 824 |
+
"grad_norm": 1.4417003393173218,
|
| 825 |
+
"learning_rate": 0.00010325581395348838,
|
| 826 |
+
"loss": 0.3601,
|
| 827 |
+
"step": 1040
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 4.883720930232558,
|
| 831 |
+
"grad_norm": 2.7006003856658936,
|
| 832 |
+
"learning_rate": 0.00010232558139534885,
|
| 833 |
+
"loss": 0.335,
|
| 834 |
+
"step": 1050
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 4.930232558139535,
|
| 838 |
+
"grad_norm": 2.294205665588379,
|
| 839 |
+
"learning_rate": 0.00010139534883720931,
|
| 840 |
+
"loss": 0.2171,
|
| 841 |
+
"step": 1060
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 4.976744186046512,
|
| 845 |
+
"grad_norm": 4.027862071990967,
|
| 846 |
+
"learning_rate": 0.00010046511627906978,
|
| 847 |
+
"loss": 0.3075,
|
| 848 |
+
"step": 1070
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 5.023255813953488,
|
| 852 |
+
"grad_norm": 2.2804319858551025,
|
| 853 |
+
"learning_rate": 9.953488372093024e-05,
|
| 854 |
+
"loss": 0.2058,
|
| 855 |
+
"step": 1080
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 5.069767441860465,
|
| 859 |
+
"grad_norm": 3.193336009979248,
|
| 860 |
+
"learning_rate": 9.86046511627907e-05,
|
| 861 |
+
"loss": 0.1606,
|
| 862 |
+
"step": 1090
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 5.116279069767442,
|
| 866 |
+
"grad_norm": 1.5344799757003784,
|
| 867 |
+
"learning_rate": 9.767441860465116e-05,
|
| 868 |
+
"loss": 0.295,
|
| 869 |
+
"step": 1100
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 5.116279069767442,
|
| 873 |
+
"eval_accuracy": 0.8811881188118812,
|
| 874 |
+
"eval_loss": 0.4204191267490387,
|
| 875 |
+
"eval_runtime": 7.9394,
|
| 876 |
+
"eval_samples_per_second": 76.329,
|
| 877 |
+
"eval_steps_per_second": 9.573,
|
| 878 |
+
"step": 1100
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 5.162790697674419,
|
| 882 |
+
"grad_norm": 2.80682110786438,
|
| 883 |
+
"learning_rate": 9.674418604651163e-05,
|
| 884 |
+
"loss": 0.2268,
|
| 885 |
+
"step": 1110
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 5.209302325581396,
|
| 889 |
+
"grad_norm": 2.198103904724121,
|
| 890 |
+
"learning_rate": 9.58139534883721e-05,
|
| 891 |
+
"loss": 0.2226,
|
| 892 |
+
"step": 1120
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 5.2558139534883725,
|
| 896 |
+
"grad_norm": 3.6225826740264893,
|
| 897 |
+
"learning_rate": 9.488372093023256e-05,
|
| 898 |
+
"loss": 0.1485,
|
| 899 |
+
"step": 1130
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 5.3023255813953485,
|
| 903 |
+
"grad_norm": 1.2400827407836914,
|
| 904 |
+
"learning_rate": 9.395348837209302e-05,
|
| 905 |
+
"loss": 0.1442,
|
| 906 |
+
"step": 1140
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 5.348837209302325,
|
| 910 |
+
"grad_norm": 1.0701277256011963,
|
| 911 |
+
"learning_rate": 9.30232558139535e-05,
|
| 912 |
+
"loss": 0.2154,
|
| 913 |
+
"step": 1150
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 5.395348837209302,
|
| 917 |
+
"grad_norm": 1.8416105508804321,
|
| 918 |
+
"learning_rate": 9.209302325581396e-05,
|
| 919 |
+
"loss": 0.2288,
|
| 920 |
+
"step": 1160
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 5.441860465116279,
|
| 924 |
+
"grad_norm": 5.05810546875,
|
| 925 |
+
"learning_rate": 9.116279069767443e-05,
|
| 926 |
+
"loss": 0.3002,
|
| 927 |
+
"step": 1170
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 5.488372093023256,
|
| 931 |
+
"grad_norm": 2.076577663421631,
|
| 932 |
+
"learning_rate": 9.023255813953489e-05,
|
| 933 |
+
"loss": 0.1431,
|
| 934 |
+
"step": 1180
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 5.534883720930233,
|
| 938 |
+
"grad_norm": 1.4769999980926514,
|
| 939 |
+
"learning_rate": 8.930232558139535e-05,
|
| 940 |
+
"loss": 0.2355,
|
| 941 |
+
"step": 1190
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 5.5813953488372094,
|
| 945 |
+
"grad_norm": 4.101127624511719,
|
| 946 |
+
"learning_rate": 8.837209302325582e-05,
|
| 947 |
+
"loss": 0.2342,
|
| 948 |
+
"step": 1200
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 5.5813953488372094,
|
| 952 |
+
"eval_accuracy": 0.8910891089108911,
|
| 953 |
+
"eval_loss": 0.3933403491973877,
|
| 954 |
+
"eval_runtime": 8.1318,
|
| 955 |
+
"eval_samples_per_second": 74.523,
|
| 956 |
+
"eval_steps_per_second": 9.346,
|
| 957 |
+
"step": 1200
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 5.627906976744186,
|
| 961 |
+
"grad_norm": 3.6835029125213623,
|
| 962 |
+
"learning_rate": 8.744186046511629e-05,
|
| 963 |
+
"loss": 0.221,
|
| 964 |
+
"step": 1210
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 5.674418604651163,
|
| 968 |
+
"grad_norm": 5.327223300933838,
|
| 969 |
+
"learning_rate": 8.651162790697674e-05,
|
| 970 |
+
"loss": 0.2887,
|
| 971 |
+
"step": 1220
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 5.720930232558139,
|
| 975 |
+
"grad_norm": 1.522955060005188,
|
| 976 |
+
"learning_rate": 8.558139534883721e-05,
|
| 977 |
+
"loss": 0.2057,
|
| 978 |
+
"step": 1230
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 5.767441860465116,
|
| 982 |
+
"grad_norm": 5.1004133224487305,
|
| 983 |
+
"learning_rate": 8.465116279069768e-05,
|
| 984 |
+
"loss": 0.2433,
|
| 985 |
+
"step": 1240
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 5.813953488372093,
|
| 989 |
+
"grad_norm": 1.7744640111923218,
|
| 990 |
+
"learning_rate": 8.372093023255814e-05,
|
| 991 |
+
"loss": 0.1493,
|
| 992 |
+
"step": 1250
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 5.8604651162790695,
|
| 996 |
+
"grad_norm": 4.374938488006592,
|
| 997 |
+
"learning_rate": 8.27906976744186e-05,
|
| 998 |
+
"loss": 0.3513,
|
| 999 |
+
"step": 1260
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 5.906976744186046,
|
| 1003 |
+
"grad_norm": 2.1579113006591797,
|
| 1004 |
+
"learning_rate": 8.186046511627907e-05,
|
| 1005 |
+
"loss": 0.1253,
|
| 1006 |
+
"step": 1270
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 5.953488372093023,
|
| 1010 |
+
"grad_norm": 3.5076637268066406,
|
| 1011 |
+
"learning_rate": 8.093023255813953e-05,
|
| 1012 |
+
"loss": 0.2042,
|
| 1013 |
+
"step": 1280
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 6.0,
|
| 1017 |
+
"grad_norm": 2.294401168823242,
|
| 1018 |
+
"learning_rate": 8e-05,
|
| 1019 |
+
"loss": 0.3253,
|
| 1020 |
+
"step": 1290
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 6.046511627906977,
|
| 1024 |
+
"grad_norm": 0.5659641027450562,
|
| 1025 |
+
"learning_rate": 7.906976744186047e-05,
|
| 1026 |
+
"loss": 0.1629,
|
| 1027 |
+
"step": 1300
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 6.046511627906977,
|
| 1031 |
+
"eval_accuracy": 0.8976897689768977,
|
| 1032 |
+
"eval_loss": 0.36343997716903687,
|
| 1033 |
+
"eval_runtime": 8.0696,
|
| 1034 |
+
"eval_samples_per_second": 75.096,
|
| 1035 |
+
"eval_steps_per_second": 9.418,
|
| 1036 |
+
"step": 1300
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 6.093023255813954,
|
| 1040 |
+
"grad_norm": 3.126523971557617,
|
| 1041 |
+
"learning_rate": 7.813953488372094e-05,
|
| 1042 |
+
"loss": 0.1177,
|
| 1043 |
+
"step": 1310
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 6.1395348837209305,
|
| 1047 |
+
"grad_norm": 7.47089958190918,
|
| 1048 |
+
"learning_rate": 7.72093023255814e-05,
|
| 1049 |
+
"loss": 0.2036,
|
| 1050 |
+
"step": 1320
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 6.186046511627907,
|
| 1054 |
+
"grad_norm": 1.5586128234863281,
|
| 1055 |
+
"learning_rate": 7.627906976744187e-05,
|
| 1056 |
+
"loss": 0.1308,
|
| 1057 |
+
"step": 1330
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 6.232558139534884,
|
| 1061 |
+
"grad_norm": 2.230147361755371,
|
| 1062 |
+
"learning_rate": 7.534883720930233e-05,
|
| 1063 |
+
"loss": 0.2078,
|
| 1064 |
+
"step": 1340
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 6.27906976744186,
|
| 1068 |
+
"grad_norm": 4.571086883544922,
|
| 1069 |
+
"learning_rate": 7.441860465116279e-05,
|
| 1070 |
+
"loss": 0.131,
|
| 1071 |
+
"step": 1350
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 6.325581395348837,
|
| 1075 |
+
"grad_norm": 0.7467085719108582,
|
| 1076 |
+
"learning_rate": 7.348837209302326e-05,
|
| 1077 |
+
"loss": 0.1465,
|
| 1078 |
+
"step": 1360
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 6.372093023255814,
|
| 1082 |
+
"grad_norm": 1.2842942476272583,
|
| 1083 |
+
"learning_rate": 7.255813953488373e-05,
|
| 1084 |
+
"loss": 0.1553,
|
| 1085 |
+
"step": 1370
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 6.4186046511627906,
|
| 1089 |
+
"grad_norm": 0.23809537291526794,
|
| 1090 |
+
"learning_rate": 7.162790697674418e-05,
|
| 1091 |
+
"loss": 0.1314,
|
| 1092 |
+
"step": 1380
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 6.465116279069767,
|
| 1096 |
+
"grad_norm": 8.466991424560547,
|
| 1097 |
+
"learning_rate": 7.069767441860465e-05,
|
| 1098 |
+
"loss": 0.2773,
|
| 1099 |
+
"step": 1390
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 6.511627906976744,
|
| 1103 |
+
"grad_norm": 11.282208442687988,
|
| 1104 |
+
"learning_rate": 6.976744186046513e-05,
|
| 1105 |
+
"loss": 0.2041,
|
| 1106 |
+
"step": 1400
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 6.511627906976744,
|
| 1110 |
+
"eval_accuracy": 0.8910891089108911,
|
| 1111 |
+
"eval_loss": 0.4006832540035248,
|
| 1112 |
+
"eval_runtime": 10.4521,
|
| 1113 |
+
"eval_samples_per_second": 57.979,
|
| 1114 |
+
"eval_steps_per_second": 7.271,
|
| 1115 |
+
"step": 1400
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 6.558139534883721,
|
| 1119 |
+
"grad_norm": 0.22673094272613525,
|
| 1120 |
+
"learning_rate": 6.883720930232558e-05,
|
| 1121 |
+
"loss": 0.1841,
|
| 1122 |
+
"step": 1410
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 6.604651162790698,
|
| 1126 |
+
"grad_norm": 1.8003208637237549,
|
| 1127 |
+
"learning_rate": 6.790697674418604e-05,
|
| 1128 |
+
"loss": 0.2147,
|
| 1129 |
+
"step": 1420
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 6.651162790697675,
|
| 1133 |
+
"grad_norm": 3.2758359909057617,
|
| 1134 |
+
"learning_rate": 6.697674418604652e-05,
|
| 1135 |
+
"loss": 0.1884,
|
| 1136 |
+
"step": 1430
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 6.6976744186046515,
|
| 1140 |
+
"grad_norm": 2.0551984310150146,
|
| 1141 |
+
"learning_rate": 6.604651162790698e-05,
|
| 1142 |
+
"loss": 0.2099,
|
| 1143 |
+
"step": 1440
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 6.7441860465116275,
|
| 1147 |
+
"grad_norm": 0.6447792053222656,
|
| 1148 |
+
"learning_rate": 6.511627906976745e-05,
|
| 1149 |
+
"loss": 0.114,
|
| 1150 |
+
"step": 1450
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 6.790697674418604,
|
| 1154 |
+
"grad_norm": 0.8234413862228394,
|
| 1155 |
+
"learning_rate": 6.418604651162791e-05,
|
| 1156 |
+
"loss": 0.141,
|
| 1157 |
+
"step": 1460
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 6.837209302325581,
|
| 1161 |
+
"grad_norm": 4.1033406257629395,
|
| 1162 |
+
"learning_rate": 6.325581395348838e-05,
|
| 1163 |
+
"loss": 0.1895,
|
| 1164 |
+
"step": 1470
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 6.883720930232558,
|
| 1168 |
+
"grad_norm": 3.889683246612549,
|
| 1169 |
+
"learning_rate": 6.232558139534884e-05,
|
| 1170 |
+
"loss": 0.1317,
|
| 1171 |
+
"step": 1480
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 6.930232558139535,
|
| 1175 |
+
"grad_norm": 4.772053241729736,
|
| 1176 |
+
"learning_rate": 6.139534883720931e-05,
|
| 1177 |
+
"loss": 0.2148,
|
| 1178 |
+
"step": 1490
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 6.976744186046512,
|
| 1182 |
+
"grad_norm": 3.037203788757324,
|
| 1183 |
+
"learning_rate": 6.0465116279069765e-05,
|
| 1184 |
+
"loss": 0.1668,
|
| 1185 |
+
"step": 1500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 6.976744186046512,
|
| 1189 |
+
"eval_accuracy": 0.8927392739273927,
|
| 1190 |
+
"eval_loss": 0.3843345642089844,
|
| 1191 |
+
"eval_runtime": 8.3737,
|
| 1192 |
+
"eval_samples_per_second": 72.37,
|
| 1193 |
+
"eval_steps_per_second": 9.076,
|
| 1194 |
+
"step": 1500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 7.023255813953488,
|
| 1198 |
+
"grad_norm": 4.356062889099121,
|
| 1199 |
+
"learning_rate": 5.953488372093024e-05,
|
| 1200 |
+
"loss": 0.2231,
|
| 1201 |
+
"step": 1510
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 7.069767441860465,
|
| 1205 |
+
"grad_norm": 3.485504388809204,
|
| 1206 |
+
"learning_rate": 5.8604651162790704e-05,
|
| 1207 |
+
"loss": 0.1391,
|
| 1208 |
+
"step": 1520
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 7.116279069767442,
|
| 1212 |
+
"grad_norm": 1.0620895624160767,
|
| 1213 |
+
"learning_rate": 5.767441860465117e-05,
|
| 1214 |
+
"loss": 0.1369,
|
| 1215 |
+
"step": 1530
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 7.162790697674419,
|
| 1219 |
+
"grad_norm": 0.46343836188316345,
|
| 1220 |
+
"learning_rate": 5.674418604651163e-05,
|
| 1221 |
+
"loss": 0.1639,
|
| 1222 |
+
"step": 1540
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 7.209302325581396,
|
| 1226 |
+
"grad_norm": 5.317925453186035,
|
| 1227 |
+
"learning_rate": 5.5813953488372095e-05,
|
| 1228 |
+
"loss": 0.1061,
|
| 1229 |
+
"step": 1550
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 7.2558139534883725,
|
| 1233 |
+
"grad_norm": 2.8796050548553467,
|
| 1234 |
+
"learning_rate": 5.488372093023256e-05,
|
| 1235 |
+
"loss": 0.164,
|
| 1236 |
+
"step": 1560
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 7.3023255813953485,
|
| 1240 |
+
"grad_norm": 2.1574888229370117,
|
| 1241 |
+
"learning_rate": 5.3953488372093034e-05,
|
| 1242 |
+
"loss": 0.0833,
|
| 1243 |
+
"step": 1570
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 7.348837209302325,
|
| 1247 |
+
"grad_norm": 4.602242469787598,
|
| 1248 |
+
"learning_rate": 5.3023255813953486e-05,
|
| 1249 |
+
"loss": 0.1002,
|
| 1250 |
+
"step": 1580
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 7.395348837209302,
|
| 1254 |
+
"grad_norm": 1.2774103879928589,
|
| 1255 |
+
"learning_rate": 5.209302325581395e-05,
|
| 1256 |
+
"loss": 0.0654,
|
| 1257 |
+
"step": 1590
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 7.441860465116279,
|
| 1261 |
+
"grad_norm": 0.37218332290649414,
|
| 1262 |
+
"learning_rate": 5.1162790697674425e-05,
|
| 1263 |
+
"loss": 0.0976,
|
| 1264 |
+
"step": 1600
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 7.441860465116279,
|
| 1268 |
+
"eval_accuracy": 0.8927392739273927,
|
| 1269 |
+
"eval_loss": 0.4061937630176544,
|
| 1270 |
+
"eval_runtime": 8.5468,
|
| 1271 |
+
"eval_samples_per_second": 70.904,
|
| 1272 |
+
"eval_steps_per_second": 8.892,
|
| 1273 |
+
"step": 1600
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 7.488372093023256,
|
| 1277 |
+
"grad_norm": 1.125712275505066,
|
| 1278 |
+
"learning_rate": 5.023255813953489e-05,
|
| 1279 |
+
"loss": 0.1522,
|
| 1280 |
+
"step": 1610
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 7.534883720930233,
|
| 1284 |
+
"grad_norm": 3.255833625793457,
|
| 1285 |
+
"learning_rate": 4.930232558139535e-05,
|
| 1286 |
+
"loss": 0.154,
|
| 1287 |
+
"step": 1620
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 7.5813953488372094,
|
| 1291 |
+
"grad_norm": 3.5783629417419434,
|
| 1292 |
+
"learning_rate": 4.8372093023255816e-05,
|
| 1293 |
+
"loss": 0.1385,
|
| 1294 |
+
"step": 1630
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 7.627906976744186,
|
| 1298 |
+
"grad_norm": 4.361517429351807,
|
| 1299 |
+
"learning_rate": 4.744186046511628e-05,
|
| 1300 |
+
"loss": 0.1105,
|
| 1301 |
+
"step": 1640
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 7.674418604651163,
|
| 1305 |
+
"grad_norm": 1.0743229389190674,
|
| 1306 |
+
"learning_rate": 4.651162790697675e-05,
|
| 1307 |
+
"loss": 0.1461,
|
| 1308 |
+
"step": 1650
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 7.720930232558139,
|
| 1312 |
+
"grad_norm": 1.155990481376648,
|
| 1313 |
+
"learning_rate": 4.5581395348837214e-05,
|
| 1314 |
+
"loss": 0.1247,
|
| 1315 |
+
"step": 1660
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 7.767441860465116,
|
| 1319 |
+
"grad_norm": 0.03133929520845413,
|
| 1320 |
+
"learning_rate": 4.465116279069767e-05,
|
| 1321 |
+
"loss": 0.0625,
|
| 1322 |
+
"step": 1670
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 7.813953488372093,
|
| 1326 |
+
"grad_norm": 1.1082956790924072,
|
| 1327 |
+
"learning_rate": 4.3720930232558146e-05,
|
| 1328 |
+
"loss": 0.2145,
|
| 1329 |
+
"step": 1680
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 7.8604651162790695,
|
| 1333 |
+
"grad_norm": 2.6899712085723877,
|
| 1334 |
+
"learning_rate": 4.2790697674418605e-05,
|
| 1335 |
+
"loss": 0.1182,
|
| 1336 |
+
"step": 1690
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 7.906976744186046,
|
| 1340 |
+
"grad_norm": 5.237185955047607,
|
| 1341 |
+
"learning_rate": 4.186046511627907e-05,
|
| 1342 |
+
"loss": 0.1275,
|
| 1343 |
+
"step": 1700
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 7.906976744186046,
|
| 1347 |
+
"eval_accuracy": 0.8894389438943895,
|
| 1348 |
+
"eval_loss": 0.3860972821712494,
|
| 1349 |
+
"eval_runtime": 8.3425,
|
| 1350 |
+
"eval_samples_per_second": 72.64,
|
| 1351 |
+
"eval_steps_per_second": 9.11,
|
| 1352 |
+
"step": 1700
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 7.953488372093023,
|
| 1356 |
+
"grad_norm": 1.522539496421814,
|
| 1357 |
+
"learning_rate": 4.093023255813954e-05,
|
| 1358 |
+
"loss": 0.1341,
|
| 1359 |
+
"step": 1710
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 8.0,
|
| 1363 |
+
"grad_norm": 0.16910937428474426,
|
| 1364 |
+
"learning_rate": 4e-05,
|
| 1365 |
+
"loss": 0.1398,
|
| 1366 |
+
"step": 1720
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 8.046511627906977,
|
| 1370 |
+
"grad_norm": 3.6039626598358154,
|
| 1371 |
+
"learning_rate": 3.906976744186047e-05,
|
| 1372 |
+
"loss": 0.089,
|
| 1373 |
+
"step": 1730
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 8.093023255813954,
|
| 1377 |
+
"grad_norm": 3.6135542392730713,
|
| 1378 |
+
"learning_rate": 3.8139534883720935e-05,
|
| 1379 |
+
"loss": 0.1332,
|
| 1380 |
+
"step": 1740
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 8.13953488372093,
|
| 1384 |
+
"grad_norm": 2.710392713546753,
|
| 1385 |
+
"learning_rate": 3.7209302325581394e-05,
|
| 1386 |
+
"loss": 0.1241,
|
| 1387 |
+
"step": 1750
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 8.186046511627907,
|
| 1391 |
+
"grad_norm": 5.7339067459106445,
|
| 1392 |
+
"learning_rate": 3.627906976744187e-05,
|
| 1393 |
+
"loss": 0.1269,
|
| 1394 |
+
"step": 1760
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 8.232558139534884,
|
| 1398 |
+
"grad_norm": 1.990607738494873,
|
| 1399 |
+
"learning_rate": 3.5348837209302326e-05,
|
| 1400 |
+
"loss": 0.1111,
|
| 1401 |
+
"step": 1770
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 8.279069767441861,
|
| 1405 |
+
"grad_norm": 0.09675276279449463,
|
| 1406 |
+
"learning_rate": 3.441860465116279e-05,
|
| 1407 |
+
"loss": 0.116,
|
| 1408 |
+
"step": 1780
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 8.325581395348838,
|
| 1412 |
+
"grad_norm": 0.04001321643590927,
|
| 1413 |
+
"learning_rate": 3.348837209302326e-05,
|
| 1414 |
+
"loss": 0.1184,
|
| 1415 |
+
"step": 1790
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 8.372093023255815,
|
| 1419 |
+
"grad_norm": 0.477737694978714,
|
| 1420 |
+
"learning_rate": 3.2558139534883724e-05,
|
| 1421 |
+
"loss": 0.1063,
|
| 1422 |
+
"step": 1800
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 8.372093023255815,
|
| 1426 |
+
"eval_accuracy": 0.8910891089108911,
|
| 1427 |
+
"eval_loss": 0.40107399225234985,
|
| 1428 |
+
"eval_runtime": 8.3287,
|
| 1429 |
+
"eval_samples_per_second": 72.761,
|
| 1430 |
+
"eval_steps_per_second": 9.125,
|
| 1431 |
+
"step": 1800
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 8.418604651162791,
|
| 1435 |
+
"grad_norm": 0.2791198790073395,
|
| 1436 |
+
"learning_rate": 3.162790697674419e-05,
|
| 1437 |
+
"loss": 0.1073,
|
| 1438 |
+
"step": 1810
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 8.465116279069768,
|
| 1442 |
+
"grad_norm": 0.03783993795514107,
|
| 1443 |
+
"learning_rate": 3.0697674418604656e-05,
|
| 1444 |
+
"loss": 0.1584,
|
| 1445 |
+
"step": 1820
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 8.511627906976745,
|
| 1449 |
+
"grad_norm": 0.7814832925796509,
|
| 1450 |
+
"learning_rate": 2.976744186046512e-05,
|
| 1451 |
+
"loss": 0.0507,
|
| 1452 |
+
"step": 1830
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 8.55813953488372,
|
| 1456 |
+
"grad_norm": 0.22078533470630646,
|
| 1457 |
+
"learning_rate": 2.8837209302325585e-05,
|
| 1458 |
+
"loss": 0.0539,
|
| 1459 |
+
"step": 1840
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 8.604651162790697,
|
| 1463 |
+
"grad_norm": 2.293942928314209,
|
| 1464 |
+
"learning_rate": 2.7906976744186048e-05,
|
| 1465 |
+
"loss": 0.1488,
|
| 1466 |
+
"step": 1850
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 8.651162790697674,
|
| 1470 |
+
"grad_norm": 0.0680939108133316,
|
| 1471 |
+
"learning_rate": 2.6976744186046517e-05,
|
| 1472 |
+
"loss": 0.071,
|
| 1473 |
+
"step": 1860
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 8.69767441860465,
|
| 1477 |
+
"grad_norm": 2.966674327850342,
|
| 1478 |
+
"learning_rate": 2.6046511627906976e-05,
|
| 1479 |
+
"loss": 0.1212,
|
| 1480 |
+
"step": 1870
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 8.744186046511627,
|
| 1484 |
+
"grad_norm": 0.7327284216880798,
|
| 1485 |
+
"learning_rate": 2.5116279069767445e-05,
|
| 1486 |
+
"loss": 0.0865,
|
| 1487 |
+
"step": 1880
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 8.790697674418604,
|
| 1491 |
+
"grad_norm": 2.343684673309326,
|
| 1492 |
+
"learning_rate": 2.4186046511627908e-05,
|
| 1493 |
+
"loss": 0.1467,
|
| 1494 |
+
"step": 1890
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 8.837209302325581,
|
| 1498 |
+
"grad_norm": 5.241259574890137,
|
| 1499 |
+
"learning_rate": 2.3255813953488374e-05,
|
| 1500 |
+
"loss": 0.1658,
|
| 1501 |
+
"step": 1900
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 8.837209302325581,
|
| 1505 |
+
"eval_accuracy": 0.9042904290429042,
|
| 1506 |
+
"eval_loss": 0.3840368390083313,
|
| 1507 |
+
"eval_runtime": 8.2921,
|
| 1508 |
+
"eval_samples_per_second": 73.081,
|
| 1509 |
+
"eval_steps_per_second": 9.165,
|
| 1510 |
+
"step": 1900
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 8.883720930232558,
|
| 1514 |
+
"grad_norm": 0.7707066535949707,
|
| 1515 |
+
"learning_rate": 2.2325581395348837e-05,
|
| 1516 |
+
"loss": 0.1274,
|
| 1517 |
+
"step": 1910
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 8.930232558139535,
|
| 1521 |
+
"grad_norm": 2.1124603748321533,
|
| 1522 |
+
"learning_rate": 2.1395348837209303e-05,
|
| 1523 |
+
"loss": 0.1397,
|
| 1524 |
+
"step": 1920
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 8.976744186046512,
|
| 1528 |
+
"grad_norm": 0.4913456439971924,
|
| 1529 |
+
"learning_rate": 2.046511627906977e-05,
|
| 1530 |
+
"loss": 0.1697,
|
| 1531 |
+
"step": 1930
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 9.023255813953488,
|
| 1535 |
+
"grad_norm": 1.9249141216278076,
|
| 1536 |
+
"learning_rate": 1.9534883720930235e-05,
|
| 1537 |
+
"loss": 0.139,
|
| 1538 |
+
"step": 1940
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 9.069767441860465,
|
| 1542 |
+
"grad_norm": 1.6021732091903687,
|
| 1543 |
+
"learning_rate": 1.8604651162790697e-05,
|
| 1544 |
+
"loss": 0.1806,
|
| 1545 |
+
"step": 1950
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 9.116279069767442,
|
| 1549 |
+
"grad_norm": 0.47562333941459656,
|
| 1550 |
+
"learning_rate": 1.7674418604651163e-05,
|
| 1551 |
+
"loss": 0.1272,
|
| 1552 |
+
"step": 1960
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 9.162790697674419,
|
| 1556 |
+
"grad_norm": 4.387588024139404,
|
| 1557 |
+
"learning_rate": 1.674418604651163e-05,
|
| 1558 |
+
"loss": 0.1565,
|
| 1559 |
+
"step": 1970
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 9.209302325581396,
|
| 1563 |
+
"grad_norm": 0.03679787367582321,
|
| 1564 |
+
"learning_rate": 1.5813953488372095e-05,
|
| 1565 |
+
"loss": 0.039,
|
| 1566 |
+
"step": 1980
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 9.255813953488373,
|
| 1570 |
+
"grad_norm": 0.04448888823390007,
|
| 1571 |
+
"learning_rate": 1.488372093023256e-05,
|
| 1572 |
+
"loss": 0.084,
|
| 1573 |
+
"step": 1990
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 9.30232558139535,
|
| 1577 |
+
"grad_norm": 0.5206286311149597,
|
| 1578 |
+
"learning_rate": 1.3953488372093024e-05,
|
| 1579 |
+
"loss": 0.1,
|
| 1580 |
+
"step": 2000
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 9.30232558139535,
|
| 1584 |
+
"eval_accuracy": 0.900990099009901,
|
| 1585 |
+
"eval_loss": 0.3872542083263397,
|
| 1586 |
+
"eval_runtime": 8.0752,
|
| 1587 |
+
"eval_samples_per_second": 75.044,
|
| 1588 |
+
"eval_steps_per_second": 9.411,
|
| 1589 |
+
"step": 2000
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 9.348837209302326,
|
| 1593 |
+
"grad_norm": 0.08173186331987381,
|
| 1594 |
+
"learning_rate": 1.3023255813953488e-05,
|
| 1595 |
+
"loss": 0.1023,
|
| 1596 |
+
"step": 2010
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 9.395348837209303,
|
| 1600 |
+
"grad_norm": 0.45058727264404297,
|
| 1601 |
+
"learning_rate": 1.2093023255813954e-05,
|
| 1602 |
+
"loss": 0.1194,
|
| 1603 |
+
"step": 2020
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 9.44186046511628,
|
| 1607 |
+
"grad_norm": 0.05082235485315323,
|
| 1608 |
+
"learning_rate": 1.1162790697674418e-05,
|
| 1609 |
+
"loss": 0.0994,
|
| 1610 |
+
"step": 2030
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 9.488372093023255,
|
| 1614 |
+
"grad_norm": 0.16433951258659363,
|
| 1615 |
+
"learning_rate": 1.0232558139534884e-05,
|
| 1616 |
+
"loss": 0.124,
|
| 1617 |
+
"step": 2040
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 9.534883720930232,
|
| 1621 |
+
"grad_norm": 3.3077304363250732,
|
| 1622 |
+
"learning_rate": 9.302325581395349e-06,
|
| 1623 |
+
"loss": 0.1311,
|
| 1624 |
+
"step": 2050
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 9.581395348837209,
|
| 1628 |
+
"grad_norm": 3.5823140144348145,
|
| 1629 |
+
"learning_rate": 8.372093023255815e-06,
|
| 1630 |
+
"loss": 0.0295,
|
| 1631 |
+
"step": 2060
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 9.627906976744185,
|
| 1635 |
+
"grad_norm": 0.6963232159614563,
|
| 1636 |
+
"learning_rate": 7.44186046511628e-06,
|
| 1637 |
+
"loss": 0.0753,
|
| 1638 |
+
"step": 2070
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 9.674418604651162,
|
| 1642 |
+
"grad_norm": 0.858066201210022,
|
| 1643 |
+
"learning_rate": 6.511627906976744e-06,
|
| 1644 |
+
"loss": 0.0945,
|
| 1645 |
+
"step": 2080
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 9.720930232558139,
|
| 1649 |
+
"grad_norm": 0.16633135080337524,
|
| 1650 |
+
"learning_rate": 5.581395348837209e-06,
|
| 1651 |
+
"loss": 0.1215,
|
| 1652 |
+
"step": 2090
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 9.767441860465116,
|
| 1656 |
+
"grad_norm": 1.5643835067749023,
|
| 1657 |
+
"learning_rate": 4.651162790697674e-06,
|
| 1658 |
+
"loss": 0.1045,
|
| 1659 |
+
"step": 2100
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 9.767441860465116,
|
| 1663 |
+
"eval_accuracy": 0.9075907590759076,
|
| 1664 |
+
"eval_loss": 0.37872469425201416,
|
| 1665 |
+
"eval_runtime": 8.0071,
|
| 1666 |
+
"eval_samples_per_second": 75.683,
|
| 1667 |
+
"eval_steps_per_second": 9.492,
|
| 1668 |
+
"step": 2100
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 9.813953488372093,
|
| 1672 |
+
"grad_norm": 2.010878324508667,
|
| 1673 |
+
"learning_rate": 3.72093023255814e-06,
|
| 1674 |
+
"loss": 0.0887,
|
| 1675 |
+
"step": 2110
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 9.86046511627907,
|
| 1679 |
+
"grad_norm": 2.305905342102051,
|
| 1680 |
+
"learning_rate": 2.7906976744186046e-06,
|
| 1681 |
+
"loss": 0.1151,
|
| 1682 |
+
"step": 2120
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 9.906976744186046,
|
| 1686 |
+
"grad_norm": 2.8573191165924072,
|
| 1687 |
+
"learning_rate": 1.86046511627907e-06,
|
| 1688 |
+
"loss": 0.1443,
|
| 1689 |
+
"step": 2130
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 9.953488372093023,
|
| 1693 |
+
"grad_norm": 3.0255908966064453,
|
| 1694 |
+
"learning_rate": 9.30232558139535e-07,
|
| 1695 |
+
"loss": 0.1765,
|
| 1696 |
+
"step": 2140
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 10.0,
|
| 1700 |
+
"grad_norm": 2.159597635269165,
|
| 1701 |
+
"learning_rate": 0.0,
|
| 1702 |
+
"loss": 0.1705,
|
| 1703 |
+
"step": 2150
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 10.0,
|
| 1707 |
+
"step": 2150,
|
| 1708 |
+
"total_flos": 2.659620429501235e+18,
|
| 1709 |
+
"train_loss": 0.3051174351919529,
|
| 1710 |
+
"train_runtime": 1549.0078,
|
| 1711 |
+
"train_samples_per_second": 22.156,
|
| 1712 |
+
"train_steps_per_second": 1.388
|
| 1713 |
}
|
| 1714 |
],
|
| 1715 |
"logging_steps": 10,
|
| 1716 |
+
"max_steps": 2150,
|
| 1717 |
"num_input_tokens_seen": 0,
|
| 1718 |
+
"num_train_epochs": 10,
|
| 1719 |
"save_steps": 100,
|
| 1720 |
+
"total_flos": 2.659620429501235e+18,
|
| 1721 |
"train_batch_size": 16,
|
| 1722 |
"trial_name": null,
|
| 1723 |
"trial_params": null
|