🍻 cheers
Browse files- README.md +5 -2
- all_results.json +9 -9
- eval_results.json +5 -5
- runs/Apr03_18-00-45_X5C922065N/events.out.tfevents.1712217271.X5C922065N.13113.2 +3 -0
- train_results.json +4 -4
- trainer_state.json +703 -314
README.md
CHANGED
|
@@ -1,6 +1,9 @@
|
|
| 1 |
---
|
| 2 |
base_model: d071696/vit-finetune-scrap
|
| 3 |
tags:
|
|
|
|
|
|
|
|
|
|
| 4 |
- generated_from_trainer
|
| 5 |
datasets:
|
| 6 |
- arrow
|
|
@@ -13,7 +16,7 @@ model-index:
|
|
| 13 |
name: Image Classification
|
| 14 |
type: image-classification
|
| 15 |
dataset:
|
| 16 |
-
name:
|
| 17 |
type: arrow
|
| 18 |
config: default
|
| 19 |
split: train
|
|
@@ -29,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 29 |
|
| 30 |
# vit-finetune-scrap
|
| 31 |
|
| 32 |
-
This model is a fine-tuned version of [d071696/vit-finetune-scrap](https://huggingface.co/d071696/vit-finetune-scrap) on the
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
- Loss: 0.3599
|
| 35 |
- Accuracy: 0.9260
|
|
|
|
| 1 |
---
|
| 2 |
base_model: d071696/vit-finetune-scrap
|
| 3 |
tags:
|
| 4 |
+
- image-classification
|
| 5 |
+
- image-feature-extraction
|
| 6 |
+
- image-to-text
|
| 7 |
- generated_from_trainer
|
| 8 |
datasets:
|
| 9 |
- arrow
|
|
|
|
| 16 |
name: Image Classification
|
| 17 |
type: image-classification
|
| 18 |
dataset:
|
| 19 |
+
name: d071696/scraps1
|
| 20 |
type: arrow
|
| 21 |
config: default
|
| 22 |
split: train
|
|
|
|
| 32 |
|
| 33 |
# vit-finetune-scrap
|
| 34 |
|
| 35 |
+
This model is a fine-tuned version of [d071696/vit-finetune-scrap](https://huggingface.co/d071696/vit-finetune-scrap) on the d071696/scraps1 dataset.
|
| 36 |
It achieves the following results on the evaluation set:
|
| 37 |
- Loss: 0.3599
|
| 38 |
- Accuracy: 0.9260
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 9.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
"total_flos": 7.703325099767808e+17,
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second":
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.9260450160771704,
|
| 4 |
+
"eval_loss": 0.3599020838737488,
|
| 5 |
+
"eval_runtime": 9.9383,
|
| 6 |
+
"eval_samples_per_second": 62.586,
|
| 7 |
+
"eval_steps_per_second": 7.848,
|
| 8 |
"total_flos": 7.703325099767808e+17,
|
| 9 |
+
"train_loss": 0.11289434264701495,
|
| 10 |
+
"train_runtime": 3405.5271,
|
| 11 |
+
"train_samples_per_second": 2.919,
|
| 12 |
+
"train_steps_per_second": 0.365
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 9.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.9260450160771704,
|
| 4 |
+
"eval_loss": 0.3599020838737488,
|
| 5 |
+
"eval_runtime": 9.9383,
|
| 6 |
+
"eval_samples_per_second": 62.586,
|
| 7 |
+
"eval_steps_per_second": 7.848
|
| 8 |
}
|
runs/Apr03_18-00-45_X5C922065N/events.out.tfevents.1712217271.X5C922065N.13113.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:76e34509a2c262075a85ede1cd4f01009894a490b5cef5fecbb8a7a82387e238
|
| 3 |
+
size 734
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
"total_flos": 7.703325099767808e+17,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
"total_flos": 7.703325099767808e+17,
|
| 4 |
+
"train_loss": 0.11289434264701495,
|
| 5 |
+
"train_runtime": 3405.5271,
|
| 6 |
+
"train_samples_per_second": 2.919,
|
| 7 |
+
"train_steps_per_second": 0.365
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,518 +1,907 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./vit-finetune-scrap/checkpoint-
|
| 4 |
"epoch": 4.0,
|
| 5 |
-
"eval_steps":
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss":
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 0.
|
| 22 |
-
"loss":
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 0.
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 0.
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate": 0.
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 0.
|
| 79 |
-
"step": 100
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.64,
|
| 83 |
-
"eval_accuracy": 0.9485530546623794,
|
| 84 |
-
"eval_loss": 0.22496841847896576,
|
| 85 |
-
"eval_runtime": 11.2945,
|
| 86 |
-
"eval_samples_per_second": 55.071,
|
| 87 |
-
"eval_steps_per_second": 6.906,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
-
"epoch": 0.
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate": 0.
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch": 0.
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate": 0.
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 0.
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate": 0.
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
-
"epoch": 0.
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate": 0.
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 0.
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate": 0.
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch":
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate": 0.
|
| 129 |
-
"loss": 0.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"epoch":
|
| 134 |
-
"grad_norm": 0.
|
| 135 |
-
"learning_rate": 0.
|
| 136 |
-
"loss": 0.
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch":
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate": 0.
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch":
|
| 148 |
-
"grad_norm": 0.
|
| 149 |
-
"learning_rate": 0.
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
-
"epoch":
|
| 155 |
-
"grad_norm": 0.
|
| 156 |
-
"learning_rate": 0.
|
| 157 |
-
"loss": 0.
|
| 158 |
-
"step": 200
|
| 159 |
-
},
|
| 160 |
-
{
|
| 161 |
-
"epoch": 1.28,
|
| 162 |
-
"eval_accuracy": 0.9372990353697749,
|
| 163 |
-
"eval_loss": 0.24667073786258698,
|
| 164 |
-
"eval_runtime": 76.2101,
|
| 165 |
-
"eval_samples_per_second": 8.162,
|
| 166 |
-
"eval_steps_per_second": 1.023,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
-
"epoch":
|
| 171 |
-
"grad_norm": 0.
|
| 172 |
-
"learning_rate": 0.
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch":
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate": 0.
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
-
"epoch":
|
| 185 |
-
"grad_norm": 0.
|
| 186 |
-
"learning_rate": 0.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate": 0.
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch":
|
| 199 |
-
"grad_norm":
|
| 200 |
-
"learning_rate": 0.
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch":
|
| 206 |
-
"grad_norm":
|
| 207 |
-
"learning_rate": 0.
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch":
|
| 213 |
-
"grad_norm": 0.
|
| 214 |
-
"learning_rate": 0.
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch":
|
| 220 |
-
"grad_norm":
|
| 221 |
-
"learning_rate": 0.
|
| 222 |
-
"loss": 0.
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
-
"epoch":
|
| 227 |
-
"grad_norm":
|
| 228 |
-
"learning_rate": 0.
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch":
|
| 234 |
-
"grad_norm": 0.
|
| 235 |
-
"learning_rate": 0.
|
| 236 |
-
"loss": 0.
|
| 237 |
-
"step": 300
|
| 238 |
-
},
|
| 239 |
-
{
|
| 240 |
-
"epoch": 1.92,
|
| 241 |
-
"eval_accuracy": 0.954983922829582,
|
| 242 |
-
"eval_loss": 0.15877817571163177,
|
| 243 |
-
"eval_runtime": 9.849,
|
| 244 |
-
"eval_samples_per_second": 63.153,
|
| 245 |
-
"eval_steps_per_second": 7.92,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch": 1.
|
| 250 |
-
"grad_norm":
|
| 251 |
-
"learning_rate": 0.
|
| 252 |
-
"loss": 0.
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
-
"epoch":
|
| 257 |
-
"grad_norm":
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch":
|
| 264 |
-
"grad_norm": 0.
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch":
|
| 271 |
-
"grad_norm": 0.
|
| 272 |
-
"learning_rate":
|
| 273 |
-
"loss": 0.
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
-
"epoch":
|
| 278 |
-
"grad_norm": 0.
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"grad_norm": 0.
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
-
"epoch":
|
| 292 |
-
"grad_norm":
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
-
"epoch":
|
| 299 |
-
"grad_norm": 0.
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch":
|
| 306 |
-
"grad_norm": 0.
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch":
|
| 313 |
-
"grad_norm":
|
| 314 |
-
"learning_rate":
|
| 315 |
-
"loss": 0.
|
| 316 |
-
"step": 400
|
| 317 |
-
},
|
| 318 |
-
{
|
| 319 |
-
"epoch": 2.56,
|
| 320 |
-
"eval_accuracy": 0.9533762057877814,
|
| 321 |
-
"eval_loss": 0.16905710101127625,
|
| 322 |
-
"eval_runtime": 9.8491,
|
| 323 |
-
"eval_samples_per_second": 63.153,
|
| 324 |
-
"eval_steps_per_second": 7.92,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
-
"epoch":
|
| 329 |
-
"grad_norm": 0.
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch":
|
| 336 |
-
"grad_norm":
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
-
"grad_norm": 0.
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch":
|
| 350 |
-
"grad_norm": 0.
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"grad_norm": 0.
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
-
"grad_norm":
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch":
|
| 371 |
-
"grad_norm": 0.
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch":
|
| 378 |
-
"grad_norm": 0.
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch":
|
| 392 |
-
"grad_norm": 0.
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch":
|
| 399 |
-
"
|
| 400 |
-
"
|
| 401 |
-
"
|
| 402 |
-
"eval_samples_per_second": 62.878,
|
| 403 |
-
"eval_steps_per_second": 7.885,
|
| 404 |
-
"step": 500
|
| 405 |
-
},
|
| 406 |
-
{
|
| 407 |
-
"epoch": 3.27,
|
| 408 |
-
"grad_norm": 0.03763527050614357,
|
| 409 |
-
"learning_rate": 3.653846153846154e-05,
|
| 410 |
-
"loss": 0.0118,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
-
"epoch":
|
| 415 |
-
"grad_norm":
|
| 416 |
-
"learning_rate":
|
| 417 |
-
"loss": 0.
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
-
"epoch":
|
| 422 |
-
"grad_norm": 0.
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch":
|
| 429 |
-
"grad_norm": 0.
|
| 430 |
-
"learning_rate":
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
-
"epoch":
|
| 436 |
-
"grad_norm": 0.
|
| 437 |
-
"learning_rate":
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
-
"epoch":
|
| 443 |
-
"grad_norm": 0.
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
-
"epoch":
|
| 450 |
-
"grad_norm": 0.
|
| 451 |
-
"learning_rate":
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch":
|
| 457 |
-
"grad_norm":
|
| 458 |
-
"learning_rate":
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
-
"epoch":
|
| 464 |
-
"grad_norm": 0.
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"grad_norm": 0.
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
-
"epoch":
|
| 478 |
-
"
|
| 479 |
-
"
|
| 480 |
-
"
|
| 481 |
-
"eval_samples_per_second": 62.899,
|
| 482 |
-
"eval_steps_per_second": 7.888,
|
| 483 |
-
"step": 600
|
| 484 |
-
},
|
| 485 |
-
{
|
| 486 |
-
"epoch": 3.91,
|
| 487 |
-
"grad_norm": 0.04359051212668419,
|
| 488 |
-
"learning_rate": 4.487179487179488e-06,
|
| 489 |
-
"loss": 0.0109,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
-
"epoch":
|
| 494 |
-
"grad_norm": 0.
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 620
|
| 498 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 499 |
{
|
| 500 |
"epoch": 4.0,
|
| 501 |
-
"step":
|
| 502 |
"total_flos": 7.703325099767808e+17,
|
| 503 |
-
"train_loss": 0.
|
| 504 |
-
"train_runtime":
|
| 505 |
-
"train_samples_per_second":
|
| 506 |
-
"train_steps_per_second":
|
| 507 |
}
|
| 508 |
],
|
| 509 |
"logging_steps": 10,
|
| 510 |
-
"max_steps":
|
| 511 |
"num_input_tokens_seen": 0,
|
| 512 |
"num_train_epochs": 4,
|
| 513 |
-
"save_steps":
|
| 514 |
"total_flos": 7.703325099767808e+17,
|
| 515 |
-
"train_batch_size":
|
| 516 |
"trial_name": null,
|
| 517 |
"trial_params": null
|
| 518 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.3599020838737488,
|
| 3 |
+
"best_model_checkpoint": "./vit-finetune-scrap/checkpoint-1000",
|
| 4 |
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 1244,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"grad_norm": 0.7885215282440186,
|
| 14 |
+
"learning_rate": 0.00019839228295819936,
|
| 15 |
+
"loss": 0.3445,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.06,
|
| 20 |
+
"grad_norm": 0.6006595492362976,
|
| 21 |
+
"learning_rate": 0.00019678456591639874,
|
| 22 |
+
"loss": 0.0458,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.1,
|
| 27 |
+
"grad_norm": 0.0847816988825798,
|
| 28 |
+
"learning_rate": 0.00019517684887459809,
|
| 29 |
+
"loss": 0.1535,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.13,
|
| 34 |
+
"grad_norm": 10.684067726135254,
|
| 35 |
+
"learning_rate": 0.00019356913183279743,
|
| 36 |
+
"loss": 0.1208,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.16,
|
| 41 |
+
"grad_norm": 0.08578234165906906,
|
| 42 |
+
"learning_rate": 0.00019196141479099678,
|
| 43 |
+
"loss": 0.0208,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.19,
|
| 48 |
+
"grad_norm": 0.06457830220460892,
|
| 49 |
+
"learning_rate": 0.00019035369774919616,
|
| 50 |
+
"loss": 0.0838,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.23,
|
| 55 |
+
"grad_norm": 12.919880867004395,
|
| 56 |
+
"learning_rate": 0.0001887459807073955,
|
| 57 |
+
"loss": 0.0491,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.26,
|
| 62 |
+
"grad_norm": 10.229437828063965,
|
| 63 |
+
"learning_rate": 0.00018713826366559486,
|
| 64 |
+
"loss": 0.1614,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.29,
|
| 69 |
+
"grad_norm": 0.03680131584405899,
|
| 70 |
+
"learning_rate": 0.0001855305466237942,
|
| 71 |
+
"loss": 0.2226,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.32,
|
| 76 |
+
"grad_norm": 1.575692892074585,
|
| 77 |
+
"learning_rate": 0.0001839228295819936,
|
| 78 |
+
"loss": 0.3464,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.35,
|
| 83 |
+
"grad_norm": 11.554421424865723,
|
| 84 |
+
"learning_rate": 0.00018231511254019294,
|
| 85 |
+
"loss": 0.267,
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.39,
|
| 90 |
+
"grad_norm": 11.113668441772461,
|
| 91 |
+
"learning_rate": 0.00018070739549839229,
|
| 92 |
+
"loss": 0.3107,
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.42,
|
| 97 |
+
"grad_norm": 5.346070766448975,
|
| 98 |
+
"learning_rate": 0.00017909967845659166,
|
| 99 |
+
"loss": 0.6457,
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.45,
|
| 104 |
+
"grad_norm": 31.744802474975586,
|
| 105 |
+
"learning_rate": 0.000177491961414791,
|
| 106 |
+
"loss": 0.4913,
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.48,
|
| 111 |
+
"grad_norm": 6.202216148376465,
|
| 112 |
+
"learning_rate": 0.00017588424437299036,
|
| 113 |
+
"loss": 0.6627,
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.51,
|
| 118 |
+
"grad_norm": 12.808435440063477,
|
| 119 |
+
"learning_rate": 0.0001742765273311897,
|
| 120 |
+
"loss": 0.4071,
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.55,
|
| 125 |
+
"grad_norm": 0.17637468874454498,
|
| 126 |
+
"learning_rate": 0.0001726688102893891,
|
| 127 |
+
"loss": 0.2312,
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.58,
|
| 132 |
+
"grad_norm": 9.431242942810059,
|
| 133 |
+
"learning_rate": 0.00017106109324758844,
|
| 134 |
+
"loss": 0.1206,
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.61,
|
| 139 |
+
"grad_norm": 0.0686856061220169,
|
| 140 |
+
"learning_rate": 0.0001694533762057878,
|
| 141 |
+
"loss": 0.2657,
|
| 142 |
"step": 190
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.64,
|
| 146 |
+
"grad_norm": 0.035174936056137085,
|
| 147 |
+
"learning_rate": 0.00016784565916398716,
|
| 148 |
+
"loss": 0.2447,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 149 |
"step": 200
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.68,
|
| 153 |
+
"grad_norm": 0.20221814513206482,
|
| 154 |
+
"learning_rate": 0.0001662379421221865,
|
| 155 |
+
"loss": 0.0932,
|
| 156 |
"step": 210
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.71,
|
| 160 |
+
"grad_norm": 22.348093032836914,
|
| 161 |
+
"learning_rate": 0.00016463022508038586,
|
| 162 |
+
"loss": 0.5622,
|
| 163 |
"step": 220
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.74,
|
| 167 |
+
"grad_norm": 0.09363432973623276,
|
| 168 |
+
"learning_rate": 0.0001630225080385852,
|
| 169 |
+
"loss": 0.1989,
|
| 170 |
"step": 230
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.77,
|
| 174 |
+
"grad_norm": 0.10725115239620209,
|
| 175 |
+
"learning_rate": 0.0001614147909967846,
|
| 176 |
+
"loss": 0.1174,
|
| 177 |
"step": 240
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.8,
|
| 181 |
+
"grad_norm": 2.2332615852355957,
|
| 182 |
+
"learning_rate": 0.00015980707395498394,
|
| 183 |
+
"loss": 0.3852,
|
| 184 |
"step": 250
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.84,
|
| 188 |
+
"grad_norm": 0.08053239434957504,
|
| 189 |
+
"learning_rate": 0.0001581993569131833,
|
| 190 |
+
"loss": 0.2247,
|
| 191 |
"step": 260
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.87,
|
| 195 |
+
"grad_norm": 0.21441006660461426,
|
| 196 |
+
"learning_rate": 0.00015659163987138264,
|
| 197 |
+
"loss": 0.2879,
|
| 198 |
"step": 270
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.9,
|
| 202 |
+
"grad_norm": 0.0631105974316597,
|
| 203 |
+
"learning_rate": 0.00015498392282958201,
|
| 204 |
+
"loss": 0.2922,
|
| 205 |
"step": 280
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.93,
|
| 209 |
+
"grad_norm": 20.96392822265625,
|
| 210 |
+
"learning_rate": 0.00015337620578778136,
|
| 211 |
+
"loss": 0.0413,
|
| 212 |
"step": 290
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.96,
|
| 216 |
+
"grad_norm": 0.4054552912712097,
|
| 217 |
+
"learning_rate": 0.0001517684887459807,
|
| 218 |
+
"loss": 0.1037,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 219 |
"step": 300
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 1.0,
|
| 223 |
+
"grad_norm": 1.8209251165390015,
|
| 224 |
+
"learning_rate": 0.0001501607717041801,
|
| 225 |
+
"loss": 0.1802,
|
| 226 |
"step": 310
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 1.03,
|
| 230 |
+
"grad_norm": 11.979750633239746,
|
| 231 |
+
"learning_rate": 0.00014855305466237944,
|
| 232 |
+
"loss": 0.1788,
|
| 233 |
"step": 320
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 1.06,
|
| 237 |
+
"grad_norm": 0.10045678168535233,
|
| 238 |
+
"learning_rate": 0.0001469453376205788,
|
| 239 |
+
"loss": 0.4287,
|
| 240 |
"step": 330
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 1.09,
|
| 244 |
+
"grad_norm": 0.051606420427560806,
|
| 245 |
+
"learning_rate": 0.00014533762057877814,
|
| 246 |
+
"loss": 0.1055,
|
| 247 |
"step": 340
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 1.13,
|
| 251 |
+
"grad_norm": 0.20983509719371796,
|
| 252 |
+
"learning_rate": 0.00014372990353697752,
|
| 253 |
+
"loss": 0.1667,
|
| 254 |
"step": 350
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 1.16,
|
| 258 |
+
"grad_norm": 0.15858705341815948,
|
| 259 |
+
"learning_rate": 0.00014212218649517686,
|
| 260 |
+
"loss": 0.1135,
|
| 261 |
"step": 360
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 1.19,
|
| 265 |
+
"grad_norm": 1.420276165008545,
|
| 266 |
+
"learning_rate": 0.00014051446945337621,
|
| 267 |
+
"loss": 0.2112,
|
| 268 |
"step": 370
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 1.22,
|
| 272 |
+
"grad_norm": 0.02672198973596096,
|
| 273 |
+
"learning_rate": 0.0001389067524115756,
|
| 274 |
+
"loss": 0.0313,
|
| 275 |
"step": 380
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 1.25,
|
| 279 |
+
"grad_norm": 0.02562582679092884,
|
| 280 |
+
"learning_rate": 0.00013729903536977494,
|
| 281 |
+
"loss": 0.1069,
|
| 282 |
"step": 390
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 1.29,
|
| 286 |
+
"grad_norm": 4.197677135467529,
|
| 287 |
+
"learning_rate": 0.0001356913183279743,
|
| 288 |
+
"loss": 0.2342,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 289 |
"step": 400
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 1.32,
|
| 293 |
+
"grad_norm": 0.04306723549962044,
|
| 294 |
+
"learning_rate": 0.00013408360128617364,
|
| 295 |
+
"loss": 0.1313,
|
| 296 |
"step": 410
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 1.35,
|
| 300 |
+
"grad_norm": 4.383702278137207,
|
| 301 |
+
"learning_rate": 0.00013247588424437302,
|
| 302 |
+
"loss": 0.2145,
|
| 303 |
"step": 420
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 1.38,
|
| 307 |
+
"grad_norm": 0.20137301087379456,
|
| 308 |
+
"learning_rate": 0.00013086816720257237,
|
| 309 |
+
"loss": 0.1514,
|
| 310 |
"step": 430
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 1.41,
|
| 314 |
+
"grad_norm": 0.020689483731985092,
|
| 315 |
+
"learning_rate": 0.00012926045016077172,
|
| 316 |
+
"loss": 0.0712,
|
| 317 |
"step": 440
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 1.45,
|
| 321 |
+
"grad_norm": 0.018355347216129303,
|
| 322 |
+
"learning_rate": 0.00012765273311897106,
|
| 323 |
+
"loss": 0.0923,
|
| 324 |
"step": 450
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 1.48,
|
| 328 |
+
"grad_norm": 17.85742950439453,
|
| 329 |
+
"learning_rate": 0.00012604501607717044,
|
| 330 |
+
"loss": 0.0779,
|
| 331 |
"step": 460
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 1.51,
|
| 335 |
+
"grad_norm": 0.05268234387040138,
|
| 336 |
+
"learning_rate": 0.0001244372990353698,
|
| 337 |
+
"loss": 0.0096,
|
| 338 |
"step": 470
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 1.54,
|
| 342 |
+
"grad_norm": 0.09329644590616226,
|
| 343 |
+
"learning_rate": 0.00012282958199356914,
|
| 344 |
+
"loss": 0.0392,
|
| 345 |
"step": 480
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 1.58,
|
| 349 |
+
"grad_norm": 8.043782234191895,
|
| 350 |
+
"learning_rate": 0.0001212218649517685,
|
| 351 |
+
"loss": 0.1322,
|
| 352 |
"step": 490
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 1.61,
|
| 356 |
+
"grad_norm": 0.016368461772799492,
|
| 357 |
+
"learning_rate": 0.00011961414790996785,
|
| 358 |
+
"loss": 0.0843,
|
| 359 |
"step": 500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 1.64,
|
| 363 |
+
"grad_norm": 13.989496231079102,
|
| 364 |
+
"learning_rate": 0.0001180064308681672,
|
| 365 |
+
"loss": 0.2665,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 366 |
"step": 510
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 1.67,
|
| 370 |
+
"grad_norm": 14.703727722167969,
|
| 371 |
+
"learning_rate": 0.00011639871382636655,
|
| 372 |
+
"loss": 0.1551,
|
| 373 |
"step": 520
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 1.7,
|
| 377 |
+
"grad_norm": 0.13277527689933777,
|
| 378 |
+
"learning_rate": 0.00011479099678456593,
|
| 379 |
+
"loss": 0.1346,
|
| 380 |
"step": 530
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 1.74,
|
| 384 |
+
"grad_norm": 0.04265744984149933,
|
| 385 |
+
"learning_rate": 0.00011318327974276528,
|
| 386 |
+
"loss": 0.1725,
|
| 387 |
"step": 540
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 1.77,
|
| 391 |
+
"grad_norm": 0.04861054942011833,
|
| 392 |
+
"learning_rate": 0.00011157556270096463,
|
| 393 |
+
"loss": 0.0696,
|
| 394 |
"step": 550
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 1.8,
|
| 398 |
+
"grad_norm": 0.015978263691067696,
|
| 399 |
+
"learning_rate": 0.00010996784565916398,
|
| 400 |
+
"loss": 0.0583,
|
| 401 |
"step": 560
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 1.83,
|
| 405 |
+
"grad_norm": 0.21797218918800354,
|
| 406 |
+
"learning_rate": 0.00010836012861736335,
|
| 407 |
+
"loss": 0.1746,
|
| 408 |
"step": 570
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 1.86,
|
| 412 |
+
"grad_norm": 16.11418342590332,
|
| 413 |
+
"learning_rate": 0.0001067524115755627,
|
| 414 |
+
"loss": 0.2571,
|
| 415 |
"step": 580
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 1.9,
|
| 419 |
+
"grad_norm": 0.025191914290189743,
|
| 420 |
+
"learning_rate": 0.00010514469453376205,
|
| 421 |
+
"loss": 0.1326,
|
| 422 |
"step": 590
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 1.93,
|
| 426 |
+
"grad_norm": 0.03328488767147064,
|
| 427 |
+
"learning_rate": 0.00010353697749196143,
|
| 428 |
+
"loss": 0.1601,
|
| 429 |
"step": 600
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 1.96,
|
| 433 |
+
"grad_norm": 0.017355144023895264,
|
| 434 |
+
"learning_rate": 0.00010192926045016078,
|
| 435 |
+
"loss": 0.1607,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 436 |
"step": 610
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 1.99,
|
| 440 |
+
"grad_norm": 0.15774357318878174,
|
| 441 |
+
"learning_rate": 0.00010032154340836013,
|
| 442 |
+
"loss": 0.0182,
|
| 443 |
"step": 620
|
| 444 |
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 2.03,
|
| 447 |
+
"grad_norm": 0.039228443056344986,
|
| 448 |
+
"learning_rate": 9.871382636655949e-05,
|
| 449 |
+
"loss": 0.0898,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 2.06,
|
| 454 |
+
"grad_norm": 27.718074798583984,
|
| 455 |
+
"learning_rate": 9.710610932475884e-05,
|
| 456 |
+
"loss": 0.1895,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 2.09,
|
| 461 |
+
"grad_norm": 0.10635074228048325,
|
| 462 |
+
"learning_rate": 9.54983922829582e-05,
|
| 463 |
+
"loss": 0.0805,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 2.12,
|
| 468 |
+
"grad_norm": 0.01466443482786417,
|
| 469 |
+
"learning_rate": 9.389067524115757e-05,
|
| 470 |
+
"loss": 0.0697,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 2.15,
|
| 475 |
+
"grad_norm": 0.037685755640268326,
|
| 476 |
+
"learning_rate": 9.228295819935692e-05,
|
| 477 |
+
"loss": 0.0031,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 2.19,
|
| 482 |
+
"grad_norm": 0.009445942007005215,
|
| 483 |
+
"learning_rate": 9.067524115755628e-05,
|
| 484 |
+
"loss": 0.0636,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 2.22,
|
| 489 |
+
"grad_norm": 0.011714440770447254,
|
| 490 |
+
"learning_rate": 8.906752411575563e-05,
|
| 491 |
+
"loss": 0.0554,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 2.25,
|
| 496 |
+
"grad_norm": 0.018049364909529686,
|
| 497 |
+
"learning_rate": 8.7459807073955e-05,
|
| 498 |
+
"loss": 0.0073,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 2.28,
|
| 503 |
+
"grad_norm": 0.013684015721082687,
|
| 504 |
+
"learning_rate": 8.585209003215434e-05,
|
| 505 |
+
"loss": 0.004,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 2.32,
|
| 510 |
+
"grad_norm": 0.0740511417388916,
|
| 511 |
+
"learning_rate": 8.42443729903537e-05,
|
| 512 |
+
"loss": 0.0724,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 2.35,
|
| 517 |
+
"grad_norm": 0.0066555822268128395,
|
| 518 |
+
"learning_rate": 8.263665594855306e-05,
|
| 519 |
+
"loss": 0.0029,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.38,
|
| 524 |
+
"grad_norm": 0.00826901663094759,
|
| 525 |
+
"learning_rate": 8.102893890675242e-05,
|
| 526 |
+
"loss": 0.0745,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 2.41,
|
| 531 |
+
"grad_norm": 0.009980367496609688,
|
| 532 |
+
"learning_rate": 7.942122186495177e-05,
|
| 533 |
+
"loss": 0.0026,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.44,
|
| 538 |
+
"grad_norm": 7.373602390289307,
|
| 539 |
+
"learning_rate": 7.781350482315113e-05,
|
| 540 |
+
"loss": 0.1221,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 2.48,
|
| 545 |
+
"grad_norm": 0.011884603649377823,
|
| 546 |
+
"learning_rate": 7.62057877813505e-05,
|
| 547 |
+
"loss": 0.0404,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 2.51,
|
| 552 |
+
"grad_norm": 28.61383628845215,
|
| 553 |
+
"learning_rate": 7.459807073954984e-05,
|
| 554 |
+
"loss": 0.0483,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 2.54,
|
| 559 |
+
"grad_norm": 0.006681604776531458,
|
| 560 |
+
"learning_rate": 7.299035369774921e-05,
|
| 561 |
+
"loss": 0.0729,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 2.57,
|
| 566 |
+
"grad_norm": 0.02536676451563835,
|
| 567 |
+
"learning_rate": 7.138263665594856e-05,
|
| 568 |
+
"loss": 0.0023,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.6,
|
| 573 |
+
"grad_norm": 0.016948092728853226,
|
| 574 |
+
"learning_rate": 6.977491961414792e-05,
|
| 575 |
+
"loss": 0.0426,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.64,
|
| 580 |
+
"grad_norm": 0.021912137046456337,
|
| 581 |
+
"learning_rate": 6.816720257234727e-05,
|
| 582 |
+
"loss": 0.064,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.67,
|
| 587 |
+
"grad_norm": 0.014551441185176373,
|
| 588 |
+
"learning_rate": 6.655948553054663e-05,
|
| 589 |
+
"loss": 0.065,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.7,
|
| 594 |
+
"grad_norm": 0.3249826431274414,
|
| 595 |
+
"learning_rate": 6.495176848874598e-05,
|
| 596 |
+
"loss": 0.1171,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.73,
|
| 601 |
+
"grad_norm": 17.302183151245117,
|
| 602 |
+
"learning_rate": 6.334405144694535e-05,
|
| 603 |
+
"loss": 0.066,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.77,
|
| 608 |
+
"grad_norm": 0.03599601984024048,
|
| 609 |
+
"learning_rate": 6.173633440514471e-05,
|
| 610 |
+
"loss": 0.0203,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.8,
|
| 615 |
+
"grad_norm": 0.011334868147969246,
|
| 616 |
+
"learning_rate": 6.012861736334405e-05,
|
| 617 |
+
"loss": 0.0188,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.83,
|
| 622 |
+
"grad_norm": 0.012830100953578949,
|
| 623 |
+
"learning_rate": 5.8520900321543414e-05,
|
| 624 |
+
"loss": 0.0028,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.86,
|
| 629 |
+
"grad_norm": 0.09628697484731674,
|
| 630 |
+
"learning_rate": 5.6913183279742764e-05,
|
| 631 |
+
"loss": 0.0857,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.89,
|
| 636 |
+
"grad_norm": 0.008330035023391247,
|
| 637 |
+
"learning_rate": 5.530546623794213e-05,
|
| 638 |
+
"loss": 0.0018,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.93,
|
| 643 |
+
"grad_norm": 0.007389857899397612,
|
| 644 |
+
"learning_rate": 5.369774919614148e-05,
|
| 645 |
+
"loss": 0.0019,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.96,
|
| 650 |
+
"grad_norm": 0.011146631091833115,
|
| 651 |
+
"learning_rate": 5.209003215434084e-05,
|
| 652 |
+
"loss": 0.104,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 2.99,
|
| 657 |
+
"grad_norm": 0.021591678261756897,
|
| 658 |
+
"learning_rate": 5.048231511254019e-05,
|
| 659 |
+
"loss": 0.0016,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 3.02,
|
| 664 |
+
"grad_norm": 0.09255488216876984,
|
| 665 |
+
"learning_rate": 4.887459807073955e-05,
|
| 666 |
+
"loss": 0.1024,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 3.05,
|
| 671 |
+
"grad_norm": 0.01300421915948391,
|
| 672 |
+
"learning_rate": 4.726688102893891e-05,
|
| 673 |
+
"loss": 0.0019,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 3.09,
|
| 678 |
+
"grad_norm": 0.009576304815709591,
|
| 679 |
+
"learning_rate": 4.5659163987138265e-05,
|
| 680 |
+
"loss": 0.0036,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 3.12,
|
| 685 |
+
"grad_norm": 0.006058481056243181,
|
| 686 |
+
"learning_rate": 4.405144694533762e-05,
|
| 687 |
+
"loss": 0.0333,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 3.15,
|
| 692 |
+
"grad_norm": 0.00888640247285366,
|
| 693 |
+
"learning_rate": 4.244372990353698e-05,
|
| 694 |
+
"loss": 0.0026,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 3.18,
|
| 699 |
+
"grad_norm": 0.008608637377619743,
|
| 700 |
+
"learning_rate": 4.083601286173634e-05,
|
| 701 |
+
"loss": 0.002,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 3.22,
|
| 706 |
+
"grad_norm": 0.017658203840255737,
|
| 707 |
+
"learning_rate": 3.92282958199357e-05,
|
| 708 |
+
"loss": 0.0021,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 3.22,
|
| 713 |
+
"eval_accuracy": 0.9260450160771704,
|
| 714 |
+
"eval_loss": 0.3599020838737488,
|
| 715 |
+
"eval_runtime": 19.6251,
|
| 716 |
+
"eval_samples_per_second": 31.694,
|
| 717 |
+
"eval_steps_per_second": 3.975,
|
| 718 |
+
"step": 1000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 3.25,
|
| 722 |
+
"grad_norm": 0.015870286151766777,
|
| 723 |
+
"learning_rate": 3.7620578778135054e-05,
|
| 724 |
+
"loss": 0.002,
|
| 725 |
+
"step": 1010
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 3.28,
|
| 729 |
+
"grad_norm": 0.006019009742885828,
|
| 730 |
+
"learning_rate": 3.601286173633441e-05,
|
| 731 |
+
"loss": 0.0021,
|
| 732 |
+
"step": 1020
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 3.31,
|
| 736 |
+
"grad_norm": 0.02200642041862011,
|
| 737 |
+
"learning_rate": 3.4405144694533766e-05,
|
| 738 |
+
"loss": 0.0484,
|
| 739 |
+
"step": 1030
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 3.34,
|
| 743 |
+
"grad_norm": 0.00513369170948863,
|
| 744 |
+
"learning_rate": 3.279742765273312e-05,
|
| 745 |
+
"loss": 0.0021,
|
| 746 |
+
"step": 1040
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 3.38,
|
| 750 |
+
"grad_norm": 0.0066593969240784645,
|
| 751 |
+
"learning_rate": 3.118971061093248e-05,
|
| 752 |
+
"loss": 0.0017,
|
| 753 |
+
"step": 1050
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 3.41,
|
| 757 |
+
"grad_norm": 0.010374795645475388,
|
| 758 |
+
"learning_rate": 2.9581993569131832e-05,
|
| 759 |
+
"loss": 0.0015,
|
| 760 |
+
"step": 1060
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 3.44,
|
| 764 |
+
"grad_norm": 0.006329710595309734,
|
| 765 |
+
"learning_rate": 2.7974276527331188e-05,
|
| 766 |
+
"loss": 0.0727,
|
| 767 |
+
"step": 1070
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 3.47,
|
| 771 |
+
"grad_norm": 0.012291769497096539,
|
| 772 |
+
"learning_rate": 2.6366559485530545e-05,
|
| 773 |
+
"loss": 0.0022,
|
| 774 |
+
"step": 1080
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 3.5,
|
| 778 |
+
"grad_norm": 0.009208714589476585,
|
| 779 |
+
"learning_rate": 2.4758842443729904e-05,
|
| 780 |
+
"loss": 0.0017,
|
| 781 |
+
"step": 1090
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 3.54,
|
| 785 |
+
"grad_norm": 0.008606062270700932,
|
| 786 |
+
"learning_rate": 2.315112540192926e-05,
|
| 787 |
+
"loss": 0.0355,
|
| 788 |
+
"step": 1100
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 3.57,
|
| 792 |
+
"grad_norm": 0.005816610064357519,
|
| 793 |
+
"learning_rate": 2.154340836012862e-05,
|
| 794 |
+
"loss": 0.0024,
|
| 795 |
+
"step": 1110
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 3.6,
|
| 799 |
+
"grad_norm": 0.016983453184366226,
|
| 800 |
+
"learning_rate": 1.9935691318327977e-05,
|
| 801 |
+
"loss": 0.0022,
|
| 802 |
+
"step": 1120
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 3.63,
|
| 806 |
+
"grad_norm": 0.013002891093492508,
|
| 807 |
+
"learning_rate": 1.8327974276527333e-05,
|
| 808 |
+
"loss": 0.0027,
|
| 809 |
+
"step": 1130
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 3.67,
|
| 813 |
+
"grad_norm": 0.004919551312923431,
|
| 814 |
+
"learning_rate": 1.672025723472669e-05,
|
| 815 |
+
"loss": 0.0038,
|
| 816 |
+
"step": 1140
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 3.7,
|
| 820 |
+
"grad_norm": 0.01140748430043459,
|
| 821 |
+
"learning_rate": 1.5112540192926044e-05,
|
| 822 |
+
"loss": 0.0018,
|
| 823 |
+
"step": 1150
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 3.73,
|
| 827 |
+
"grad_norm": 0.0061045498587191105,
|
| 828 |
+
"learning_rate": 1.3504823151125404e-05,
|
| 829 |
+
"loss": 0.0015,
|
| 830 |
+
"step": 1160
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 3.76,
|
| 834 |
+
"grad_norm": 0.01204584538936615,
|
| 835 |
+
"learning_rate": 1.189710610932476e-05,
|
| 836 |
+
"loss": 0.0019,
|
| 837 |
+
"step": 1170
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 3.79,
|
| 841 |
+
"grad_norm": 0.005698794033378363,
|
| 842 |
+
"learning_rate": 1.0289389067524116e-05,
|
| 843 |
+
"loss": 0.0015,
|
| 844 |
+
"step": 1180
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 3.83,
|
| 848 |
+
"grad_norm": 0.008717156946659088,
|
| 849 |
+
"learning_rate": 8.681672025723474e-06,
|
| 850 |
+
"loss": 0.0018,
|
| 851 |
+
"step": 1190
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 3.86,
|
| 855 |
+
"grad_norm": 0.03146166726946831,
|
| 856 |
+
"learning_rate": 7.07395498392283e-06,
|
| 857 |
+
"loss": 0.0305,
|
| 858 |
+
"step": 1200
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 3.89,
|
| 862 |
+
"grad_norm": 0.0084818284958601,
|
| 863 |
+
"learning_rate": 5.466237942122187e-06,
|
| 864 |
+
"loss": 0.0046,
|
| 865 |
+
"step": 1210
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 3.92,
|
| 869 |
+
"grad_norm": 0.026195811107754707,
|
| 870 |
+
"learning_rate": 3.858520900321544e-06,
|
| 871 |
+
"loss": 0.0018,
|
| 872 |
+
"step": 1220
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 3.95,
|
| 876 |
+
"grad_norm": 0.01475218404084444,
|
| 877 |
+
"learning_rate": 2.2508038585209006e-06,
|
| 878 |
+
"loss": 0.0019,
|
| 879 |
+
"step": 1230
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 3.99,
|
| 883 |
+
"grad_norm": 0.006672242656350136,
|
| 884 |
+
"learning_rate": 6.430868167202573e-07,
|
| 885 |
+
"loss": 0.0516,
|
| 886 |
+
"step": 1240
|
| 887 |
+
},
|
| 888 |
{
|
| 889 |
"epoch": 4.0,
|
| 890 |
+
"step": 1244,
|
| 891 |
"total_flos": 7.703325099767808e+17,
|
| 892 |
+
"train_loss": 0.11289434264701495,
|
| 893 |
+
"train_runtime": 3405.5271,
|
| 894 |
+
"train_samples_per_second": 2.919,
|
| 895 |
+
"train_steps_per_second": 0.365
|
| 896 |
}
|
| 897 |
],
|
| 898 |
"logging_steps": 10,
|
| 899 |
+
"max_steps": 1244,
|
| 900 |
"num_input_tokens_seen": 0,
|
| 901 |
"num_train_epochs": 4,
|
| 902 |
+
"save_steps": 1000,
|
| 903 |
"total_flos": 7.703325099767808e+17,
|
| 904 |
+
"train_batch_size": 8,
|
| 905 |
"trial_name": null,
|
| 906 |
"trial_params": null
|
| 907 |
}
|