🍻 cheers
Browse files- README.md +6 -5
- all_results.json +10 -10
- eval_results.json +5 -5
- runs/Mar23_12-27-13_65433f580760/events.out.tfevents.1711199781.65433f580760.3628.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +1697 -231
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 |
datasets:
|
| 7 |
- renovation
|
|
@@ -14,7 +15,7 @@ model-index:
|
|
| 14 |
name: Image Classification
|
| 15 |
type: image-classification
|
| 16 |
dataset:
|
| 17 |
-
name:
|
| 18 |
type: renovation
|
| 19 |
config: default
|
| 20 |
split: validation
|
|
@@ -22,7 +23,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -30,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# vit-base-beans-demo-v5
|
| 32 |
|
| 33 |
-
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
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss:
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- renovation
|
|
|
|
| 15 |
name: Image Classification
|
| 16 |
type: image-classification
|
| 17 |
dataset:
|
| 18 |
+
name: beans
|
| 19 |
type: renovation
|
| 20 |
config: default
|
| 21 |
split: validation
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.6695059625212947
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
# vit-base-beans-demo-v5
|
| 33 |
|
| 34 |
+
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 beans dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.8460
|
| 37 |
+
- Accuracy: 0.6695
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
-
"total_flos":
|
| 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": 0.6695059625212947,
|
| 4 |
+
"eval_loss": 0.8459659218788147,
|
| 5 |
+
"eval_runtime": 36.9315,
|
| 6 |
+
"eval_samples_per_second": 31.789,
|
| 7 |
+
"eval_steps_per_second": 3.98,
|
| 8 |
+
"total_flos": 2.910419581971751e+18,
|
| 9 |
+
"train_loss": 0.4888155373286145,
|
| 10 |
+
"train_runtime": 2894.9609,
|
| 11 |
+
"train_samples_per_second": 12.973,
|
| 12 |
+
"train_steps_per_second": 0.811
|
| 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":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.6695059625212947,
|
| 4 |
+
"eval_loss": 0.8459659218788147,
|
| 5 |
+
"eval_runtime": 36.9315,
|
| 6 |
+
"eval_samples_per_second": 31.789,
|
| 7 |
+
"eval_steps_per_second": 3.98
|
| 8 |
}
|
runs/Mar23_12-27-13_65433f580760/events.out.tfevents.1711199781.65433f580760.3628.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:302b161eab291b644cbc90f82eeed5595548682cc98dcc111575d766a1cc0332
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"total_flos":
|
| 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": 2.910419581971751e+18,
|
| 4 |
+
"train_loss": 0.4888155373286145,
|
| 5 |
+
"train_runtime": 2894.9609,
|
| 6 |
+
"train_samples_per_second": 12.973,
|
| 7 |
+
"train_steps_per_second": 0.811
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,408 +1,1874 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-
|
| 4 |
"epoch": 4.0,
|
| 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.
|
| 13 |
-
"grad_norm": 1.
|
| 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":
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss":
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss":
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss":
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 0.
|
| 57 |
-
"loss":
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm": 2.
|
| 63 |
-
"learning_rate": 0.
|
| 64 |
-
"loss":
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss":
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss":
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_loss":
|
| 85 |
-
"eval_runtime":
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second": 3.
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
-
"epoch": 0.
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate": 0.
|
| 94 |
-
"loss":
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch": 0.
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate": 0.
|
| 101 |
-
"loss":
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch":
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate": 0.
|
| 108 |
-
"loss":
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
-
"epoch":
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate": 0.
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch":
|
| 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":
|
| 135 |
-
"learning_rate": 0.
|
| 136 |
-
"loss":
|
| 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":
|
| 149 |
-
"learning_rate": 0.
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
-
"epoch":
|
| 155 |
-
"grad_norm":
|
| 156 |
-
"learning_rate": 0.
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch":
|
| 162 |
-
"eval_accuracy": 0.
|
| 163 |
-
"eval_loss": 0.
|
| 164 |
-
"eval_runtime":
|
| 165 |
-
"eval_samples_per_second":
|
| 166 |
-
"eval_steps_per_second": 3.
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
-
"epoch":
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate": 0.
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch":
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate": 0.
|
| 180 |
-
"loss":
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
-
"epoch":
|
| 185 |
-
"grad_norm":
|
| 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":
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch":
|
| 206 |
-
"grad_norm":
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch":
|
| 213 |
-
"grad_norm":
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss":
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch":
|
| 220 |
-
"grad_norm":
|
| 221 |
-
"learning_rate":
|
| 222 |
-
"loss":
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
-
"epoch":
|
| 227 |
-
"grad_norm": 1.
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch":
|
| 234 |
-
"grad_norm":
|
| 235 |
-
"learning_rate":
|
| 236 |
-
"loss":
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch":
|
| 241 |
-
"eval_accuracy": 0.
|
| 242 |
-
"eval_loss":
|
| 243 |
-
"eval_runtime":
|
| 244 |
-
"eval_samples_per_second":
|
| 245 |
-
"eval_steps_per_second": 3.
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch":
|
| 250 |
-
"grad_norm":
|
| 251 |
-
"learning_rate":
|
| 252 |
-
"loss":
|
| 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":
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch":
|
| 271 |
-
"grad_norm":
|
| 272 |
-
"learning_rate":
|
| 273 |
-
"loss": 0.
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
-
"epoch":
|
| 278 |
-
"grad_norm":
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"grad_norm": 1.
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss":
|
| 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":
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch":
|
| 306 |
-
"grad_norm":
|
| 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":
|
| 320 |
-
"eval_accuracy": 0.
|
| 321 |
-
"eval_loss":
|
| 322 |
-
"eval_runtime":
|
| 323 |
-
"eval_samples_per_second": 30.
|
| 324 |
-
"eval_steps_per_second": 3.
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
-
"epoch":
|
| 329 |
-
"grad_norm":
|
| 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":
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch":
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"grad_norm":
|
| 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": 1.
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch":
|
| 378 |
-
"grad_norm":
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 490
|
| 389 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 390 |
{
|
| 391 |
"epoch": 4.0,
|
| 392 |
-
"step":
|
| 393 |
-
"total_flos":
|
| 394 |
-
"train_loss": 0.
|
| 395 |
-
"train_runtime":
|
| 396 |
-
"train_samples_per_second":
|
| 397 |
-
"train_steps_per_second":
|
| 398 |
}
|
| 399 |
],
|
| 400 |
"logging_steps": 10,
|
| 401 |
-
"max_steps":
|
| 402 |
"num_input_tokens_seen": 0,
|
| 403 |
"num_train_epochs": 4,
|
| 404 |
"save_steps": 100,
|
| 405 |
-
"total_flos":
|
| 406 |
"train_batch_size": 16,
|
| 407 |
"trial_name": null,
|
| 408 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8459659218788147,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-900",
|
| 4 |
"epoch": 4.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 2348,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 1.6497292518615723,
|
| 14 |
+
"learning_rate": 0.00019914821124361162,
|
| 15 |
+
"loss": 1.6003,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.03,
|
| 20 |
+
"grad_norm": 1.307145357131958,
|
| 21 |
+
"learning_rate": 0.00019829642248722317,
|
| 22 |
+
"loss": 1.2767,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.05,
|
| 27 |
+
"grad_norm": 2.9354941844940186,
|
| 28 |
+
"learning_rate": 0.00019744463373083478,
|
| 29 |
+
"loss": 1.2612,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.07,
|
| 34 |
+
"grad_norm": 1.3261815309524536,
|
| 35 |
+
"learning_rate": 0.00019659284497444633,
|
| 36 |
+
"loss": 1.2354,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.09,
|
| 41 |
+
"grad_norm": 1.5586915016174316,
|
| 42 |
+
"learning_rate": 0.00019574105621805794,
|
| 43 |
+
"loss": 1.0959,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.1,
|
| 48 |
+
"grad_norm": 1.490173578262329,
|
| 49 |
+
"learning_rate": 0.00019488926746166952,
|
| 50 |
+
"loss": 1.0438,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.12,
|
| 55 |
+
"grad_norm": 2.0831446647644043,
|
| 56 |
+
"learning_rate": 0.0001940374787052811,
|
| 57 |
+
"loss": 1.0841,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.14,
|
| 62 |
+
"grad_norm": 2.6207799911499023,
|
| 63 |
+
"learning_rate": 0.00019318568994889268,
|
| 64 |
+
"loss": 1.0983,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.15,
|
| 69 |
+
"grad_norm": 1.7383110523223877,
|
| 70 |
+
"learning_rate": 0.00019233390119250426,
|
| 71 |
+
"loss": 1.1775,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.17,
|
| 76 |
+
"grad_norm": 2.1954941749572754,
|
| 77 |
+
"learning_rate": 0.00019148211243611585,
|
| 78 |
+
"loss": 1.0616,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.17,
|
| 83 |
+
"eval_accuracy": 0.5817717206132879,
|
| 84 |
+
"eval_loss": 1.0267014503479004,
|
| 85 |
+
"eval_runtime": 39.3874,
|
| 86 |
+
"eval_samples_per_second": 29.806,
|
| 87 |
+
"eval_steps_per_second": 3.732,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 0.19,
|
| 92 |
+
"grad_norm": 1.597124695777893,
|
| 93 |
+
"learning_rate": 0.00019063032367972745,
|
| 94 |
+
"loss": 1.007,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 0.2,
|
| 99 |
+
"grad_norm": 1.289490818977356,
|
| 100 |
+
"learning_rate": 0.000189778534923339,
|
| 101 |
+
"loss": 1.0065,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 0.22,
|
| 106 |
+
"grad_norm": 1.7088607549667358,
|
| 107 |
+
"learning_rate": 0.00018892674616695061,
|
| 108 |
+
"loss": 1.0204,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 0.24,
|
| 113 |
+
"grad_norm": 2.730241537094116,
|
| 114 |
+
"learning_rate": 0.00018807495741056217,
|
| 115 |
+
"loss": 0.8969,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.26,
|
| 120 |
+
"grad_norm": 2.9691402912139893,
|
| 121 |
+
"learning_rate": 0.00018722316865417378,
|
| 122 |
+
"loss": 0.953,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 0.27,
|
| 127 |
+
"grad_norm": 2.2519712448120117,
|
| 128 |
+
"learning_rate": 0.00018637137989778536,
|
| 129 |
+
"loss": 0.9269,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 0.29,
|
| 134 |
+
"grad_norm": 1.8000602722167969,
|
| 135 |
+
"learning_rate": 0.00018551959114139694,
|
| 136 |
+
"loss": 1.1314,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 0.31,
|
| 141 |
+
"grad_norm": 1.5348334312438965,
|
| 142 |
+
"learning_rate": 0.00018466780238500855,
|
| 143 |
+
"loss": 0.9615,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 0.32,
|
| 148 |
+
"grad_norm": 1.599938988685608,
|
| 149 |
+
"learning_rate": 0.0001838160136286201,
|
| 150 |
+
"loss": 0.8033,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
+
"epoch": 0.34,
|
| 155 |
+
"grad_norm": 1.50412917137146,
|
| 156 |
+
"learning_rate": 0.0001829642248722317,
|
| 157 |
+
"loss": 0.9594,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.34,
|
| 162 |
+
"eval_accuracy": 0.6073253833049403,
|
| 163 |
+
"eval_loss": 0.9467767477035522,
|
| 164 |
+
"eval_runtime": 38.8829,
|
| 165 |
+
"eval_samples_per_second": 30.193,
|
| 166 |
+
"eval_steps_per_second": 3.781,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 0.36,
|
| 171 |
+
"grad_norm": 2.1896722316741943,
|
| 172 |
+
"learning_rate": 0.0001821124361158433,
|
| 173 |
+
"loss": 0.9217,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 0.37,
|
| 178 |
+
"grad_norm": 1.9687891006469727,
|
| 179 |
+
"learning_rate": 0.00018126064735945487,
|
| 180 |
+
"loss": 1.0296,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 0.39,
|
| 185 |
+
"grad_norm": 1.9628914594650269,
|
| 186 |
+
"learning_rate": 0.00018040885860306645,
|
| 187 |
+
"loss": 0.8122,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.41,
|
| 192 |
+
"grad_norm": 2.598545789718628,
|
| 193 |
+
"learning_rate": 0.00017955706984667803,
|
| 194 |
+
"loss": 0.8393,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 0.43,
|
| 199 |
+
"grad_norm": 2.2483532428741455,
|
| 200 |
+
"learning_rate": 0.0001787052810902896,
|
| 201 |
+
"loss": 0.9047,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 0.44,
|
| 206 |
+
"grad_norm": 2.1274337768554688,
|
| 207 |
+
"learning_rate": 0.0001778534923339012,
|
| 208 |
+
"loss": 0.91,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 0.46,
|
| 213 |
+
"grad_norm": 2.436018466949463,
|
| 214 |
+
"learning_rate": 0.00017700170357751277,
|
| 215 |
+
"loss": 1.0615,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 0.48,
|
| 220 |
+
"grad_norm": 2.069586992263794,
|
| 221 |
+
"learning_rate": 0.00017614991482112438,
|
| 222 |
+
"loss": 1.0799,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 0.49,
|
| 227 |
+
"grad_norm": 1.7266385555267334,
|
| 228 |
+
"learning_rate": 0.00017529812606473594,
|
| 229 |
+
"loss": 0.9465,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 0.51,
|
| 234 |
+
"grad_norm": 2.0491390228271484,
|
| 235 |
+
"learning_rate": 0.00017444633730834754,
|
| 236 |
+
"loss": 1.1785,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 0.51,
|
| 241 |
+
"eval_accuracy": 0.5868824531516184,
|
| 242 |
+
"eval_loss": 0.997596025466919,
|
| 243 |
+
"eval_runtime": 39.3421,
|
| 244 |
+
"eval_samples_per_second": 29.841,
|
| 245 |
+
"eval_steps_per_second": 3.736,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 0.53,
|
| 250 |
+
"grad_norm": 1.4697805643081665,
|
| 251 |
+
"learning_rate": 0.00017359454855195912,
|
| 252 |
+
"loss": 1.094,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 0.55,
|
| 257 |
+
"grad_norm": 2.369339942932129,
|
| 258 |
+
"learning_rate": 0.0001727427597955707,
|
| 259 |
+
"loss": 0.9398,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 0.56,
|
| 264 |
+
"grad_norm": 2.325148344039917,
|
| 265 |
+
"learning_rate": 0.00017189097103918229,
|
| 266 |
+
"loss": 0.9718,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 0.58,
|
| 271 |
+
"grad_norm": 1.9404678344726562,
|
| 272 |
+
"learning_rate": 0.00017103918228279387,
|
| 273 |
+
"loss": 0.9091,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 0.6,
|
| 278 |
+
"grad_norm": 2.4493370056152344,
|
| 279 |
+
"learning_rate": 0.00017018739352640547,
|
| 280 |
+
"loss": 0.9295,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 0.61,
|
| 285 |
+
"grad_norm": 1.6286579370498657,
|
| 286 |
+
"learning_rate": 0.00016933560477001706,
|
| 287 |
+
"loss": 1.1049,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 0.63,
|
| 292 |
+
"grad_norm": 3.559056043624878,
|
| 293 |
+
"learning_rate": 0.00016848381601362864,
|
| 294 |
+
"loss": 0.9566,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
+
"epoch": 0.65,
|
| 299 |
+
"grad_norm": 1.4250924587249756,
|
| 300 |
+
"learning_rate": 0.00016763202725724022,
|
| 301 |
+
"loss": 0.7772,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 0.66,
|
| 306 |
+
"grad_norm": 1.5668089389801025,
|
| 307 |
+
"learning_rate": 0.0001667802385008518,
|
| 308 |
+
"loss": 0.8869,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 0.68,
|
| 313 |
+
"grad_norm": 2.725231885910034,
|
| 314 |
+
"learning_rate": 0.00016592844974446338,
|
| 315 |
+
"loss": 0.865,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
+
"epoch": 0.68,
|
| 320 |
+
"eval_accuracy": 0.6388415672913118,
|
| 321 |
+
"eval_loss": 0.9287859201431274,
|
| 322 |
+
"eval_runtime": 38.5489,
|
| 323 |
+
"eval_samples_per_second": 30.455,
|
| 324 |
+
"eval_steps_per_second": 3.813,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 0.7,
|
| 329 |
+
"grad_norm": 2.6907713413238525,
|
| 330 |
+
"learning_rate": 0.00016507666098807496,
|
| 331 |
+
"loss": 0.899,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 0.72,
|
| 336 |
+
"grad_norm": 2.402860164642334,
|
| 337 |
+
"learning_rate": 0.00016422487223168654,
|
| 338 |
+
"loss": 0.9506,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 0.73,
|
| 343 |
+
"grad_norm": 2.749433994293213,
|
| 344 |
+
"learning_rate": 0.00016337308347529815,
|
| 345 |
+
"loss": 0.8529,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
+
"epoch": 0.75,
|
| 350 |
+
"grad_norm": 1.92979097366333,
|
| 351 |
+
"learning_rate": 0.0001625212947189097,
|
| 352 |
+
"loss": 0.8695,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 0.77,
|
| 357 |
+
"grad_norm": 2.793747901916504,
|
| 358 |
+
"learning_rate": 0.0001616695059625213,
|
| 359 |
+
"loss": 0.8614,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 0.78,
|
| 364 |
+
"grad_norm": 2.483780860900879,
|
| 365 |
+
"learning_rate": 0.0001608177172061329,
|
| 366 |
+
"loss": 0.9176,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 0.8,
|
| 371 |
+
"grad_norm": 1.7278929948806763,
|
| 372 |
+
"learning_rate": 0.00015996592844974447,
|
| 373 |
+
"loss": 0.9656,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 0.82,
|
| 378 |
+
"grad_norm": 2.649017810821533,
|
| 379 |
+
"learning_rate": 0.00015911413969335605,
|
| 380 |
+
"loss": 0.8653,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 0.83,
|
| 385 |
+
"grad_norm": 1.8457053899765015,
|
| 386 |
+
"learning_rate": 0.00015826235093696763,
|
| 387 |
+
"loss": 0.7707,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.85,
|
| 392 |
+
"grad_norm": 2.824699640274048,
|
| 393 |
+
"learning_rate": 0.00015741056218057921,
|
| 394 |
+
"loss": 0.8494,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.85,
|
| 399 |
+
"eval_accuracy": 0.651618398637138,
|
| 400 |
+
"eval_loss": 0.8572959303855896,
|
| 401 |
+
"eval_runtime": 38.0883,
|
| 402 |
+
"eval_samples_per_second": 30.823,
|
| 403 |
+
"eval_steps_per_second": 3.859,
|
| 404 |
+
"step": 500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.87,
|
| 408 |
+
"grad_norm": 1.9104124307632446,
|
| 409 |
+
"learning_rate": 0.0001565587734241908,
|
| 410 |
+
"loss": 0.8113,
|
| 411 |
+
"step": 510
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 0.89,
|
| 415 |
+
"grad_norm": 2.2717394828796387,
|
| 416 |
+
"learning_rate": 0.0001557069846678024,
|
| 417 |
+
"loss": 0.9194,
|
| 418 |
+
"step": 520
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.9,
|
| 422 |
+
"grad_norm": 1.891735553741455,
|
| 423 |
+
"learning_rate": 0.00015485519591141398,
|
| 424 |
+
"loss": 0.9337,
|
| 425 |
+
"step": 530
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.92,
|
| 429 |
+
"grad_norm": 4.8229146003723145,
|
| 430 |
+
"learning_rate": 0.00015400340715502557,
|
| 431 |
+
"loss": 0.9033,
|
| 432 |
+
"step": 540
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.94,
|
| 436 |
+
"grad_norm": 2.656970977783203,
|
| 437 |
+
"learning_rate": 0.00015315161839863715,
|
| 438 |
+
"loss": 0.8985,
|
| 439 |
+
"step": 550
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.95,
|
| 443 |
+
"grad_norm": 2.2908411026000977,
|
| 444 |
+
"learning_rate": 0.00015229982964224873,
|
| 445 |
+
"loss": 0.8708,
|
| 446 |
+
"step": 560
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.97,
|
| 450 |
+
"grad_norm": 2.141950845718384,
|
| 451 |
+
"learning_rate": 0.0001514480408858603,
|
| 452 |
+
"loss": 0.9298,
|
| 453 |
+
"step": 570
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.99,
|
| 457 |
+
"grad_norm": 2.5572831630706787,
|
| 458 |
+
"learning_rate": 0.00015059625212947192,
|
| 459 |
+
"loss": 0.9101,
|
| 460 |
+
"step": 580
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.01,
|
| 464 |
+
"grad_norm": 2.2453222274780273,
|
| 465 |
+
"learning_rate": 0.00014982964224872234,
|
| 466 |
+
"loss": 0.8034,
|
| 467 |
+
"step": 590
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.02,
|
| 471 |
+
"grad_norm": 2.2874865531921387,
|
| 472 |
+
"learning_rate": 0.00014897785349233392,
|
| 473 |
+
"loss": 0.8151,
|
| 474 |
+
"step": 600
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.02,
|
| 478 |
+
"eval_accuracy": 0.6396933560477002,
|
| 479 |
+
"eval_loss": 0.87294602394104,
|
| 480 |
+
"eval_runtime": 38.7251,
|
| 481 |
+
"eval_samples_per_second": 30.316,
|
| 482 |
+
"eval_steps_per_second": 3.796,
|
| 483 |
+
"step": 600
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.04,
|
| 487 |
+
"grad_norm": 2.950303554534912,
|
| 488 |
+
"learning_rate": 0.0001481260647359455,
|
| 489 |
+
"loss": 0.7484,
|
| 490 |
+
"step": 610
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.06,
|
| 494 |
+
"grad_norm": 1.9773017168045044,
|
| 495 |
+
"learning_rate": 0.00014727427597955708,
|
| 496 |
+
"loss": 0.6572,
|
| 497 |
+
"step": 620
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.07,
|
| 501 |
+
"grad_norm": 2.9777700901031494,
|
| 502 |
+
"learning_rate": 0.00014642248722316866,
|
| 503 |
+
"loss": 0.6927,
|
| 504 |
+
"step": 630
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.09,
|
| 508 |
+
"grad_norm": 3.323662519454956,
|
| 509 |
+
"learning_rate": 0.00014557069846678024,
|
| 510 |
+
"loss": 0.5812,
|
| 511 |
+
"step": 640
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.11,
|
| 515 |
+
"grad_norm": 1.9647018909454346,
|
| 516 |
+
"learning_rate": 0.00014471890971039185,
|
| 517 |
+
"loss": 0.6166,
|
| 518 |
+
"step": 650
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.12,
|
| 522 |
+
"grad_norm": 3.215794563293457,
|
| 523 |
+
"learning_rate": 0.0001438671209540034,
|
| 524 |
+
"loss": 0.602,
|
| 525 |
+
"step": 660
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.14,
|
| 529 |
+
"grad_norm": 2.8758130073547363,
|
| 530 |
+
"learning_rate": 0.000143015332197615,
|
| 531 |
+
"loss": 0.5224,
|
| 532 |
+
"step": 670
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.16,
|
| 536 |
+
"grad_norm": 2.142829179763794,
|
| 537 |
+
"learning_rate": 0.00014216354344122656,
|
| 538 |
+
"loss": 0.5663,
|
| 539 |
+
"step": 680
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.18,
|
| 543 |
+
"grad_norm": 6.860159397125244,
|
| 544 |
+
"learning_rate": 0.00014131175468483817,
|
| 545 |
+
"loss": 0.6479,
|
| 546 |
+
"step": 690
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.19,
|
| 550 |
+
"grad_norm": 3.3176701068878174,
|
| 551 |
+
"learning_rate": 0.00014045996592844975,
|
| 552 |
+
"loss": 0.5787,
|
| 553 |
+
"step": 700
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.19,
|
| 557 |
+
"eval_accuracy": 0.6448040885860307,
|
| 558 |
+
"eval_loss": 0.9067147970199585,
|
| 559 |
+
"eval_runtime": 38.2427,
|
| 560 |
+
"eval_samples_per_second": 30.699,
|
| 561 |
+
"eval_steps_per_second": 3.844,
|
| 562 |
+
"step": 700
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.21,
|
| 566 |
+
"grad_norm": 2.322371482849121,
|
| 567 |
+
"learning_rate": 0.00013960817717206133,
|
| 568 |
+
"loss": 0.6849,
|
| 569 |
+
"step": 710
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.23,
|
| 573 |
+
"grad_norm": 1.875775933265686,
|
| 574 |
+
"learning_rate": 0.00013875638841567291,
|
| 575 |
+
"loss": 0.6399,
|
| 576 |
+
"step": 720
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.24,
|
| 580 |
+
"grad_norm": 2.0012145042419434,
|
| 581 |
+
"learning_rate": 0.0001379045996592845,
|
| 582 |
+
"loss": 0.725,
|
| 583 |
+
"step": 730
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.26,
|
| 587 |
+
"grad_norm": 2.5320353507995605,
|
| 588 |
+
"learning_rate": 0.00013705281090289608,
|
| 589 |
+
"loss": 0.5306,
|
| 590 |
+
"step": 740
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.28,
|
| 594 |
+
"grad_norm": 2.29856538772583,
|
| 595 |
+
"learning_rate": 0.00013620102214650768,
|
| 596 |
+
"loss": 0.5731,
|
| 597 |
+
"step": 750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.29,
|
| 601 |
+
"grad_norm": 1.8604925870895386,
|
| 602 |
+
"learning_rate": 0.00013534923339011926,
|
| 603 |
+
"loss": 0.6806,
|
| 604 |
+
"step": 760
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.31,
|
| 608 |
+
"grad_norm": 2.6868739128112793,
|
| 609 |
+
"learning_rate": 0.00013449744463373084,
|
| 610 |
+
"loss": 0.5944,
|
| 611 |
+
"step": 770
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.33,
|
| 615 |
+
"grad_norm": 3.3680803775787354,
|
| 616 |
+
"learning_rate": 0.00013364565587734243,
|
| 617 |
+
"loss": 0.6412,
|
| 618 |
+
"step": 780
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.35,
|
| 622 |
+
"grad_norm": 2.798149824142456,
|
| 623 |
+
"learning_rate": 0.000132793867120954,
|
| 624 |
+
"loss": 0.5235,
|
| 625 |
+
"step": 790
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.36,
|
| 629 |
+
"grad_norm": 2.4862072467803955,
|
| 630 |
+
"learning_rate": 0.00013194207836456561,
|
| 631 |
+
"loss": 0.7768,
|
| 632 |
+
"step": 800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.36,
|
| 636 |
+
"eval_accuracy": 0.6533219761499148,
|
| 637 |
+
"eval_loss": 0.8995758295059204,
|
| 638 |
+
"eval_runtime": 38.4107,
|
| 639 |
+
"eval_samples_per_second": 30.564,
|
| 640 |
+
"eval_steps_per_second": 3.827,
|
| 641 |
+
"step": 800
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.38,
|
| 645 |
+
"grad_norm": 3.291276216506958,
|
| 646 |
+
"learning_rate": 0.00013109028960817717,
|
| 647 |
+
"loss": 0.669,
|
| 648 |
+
"step": 810
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.4,
|
| 652 |
+
"grad_norm": 2.814397096633911,
|
| 653 |
+
"learning_rate": 0.00013023850085178878,
|
| 654 |
+
"loss": 0.5539,
|
| 655 |
+
"step": 820
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.41,
|
| 659 |
+
"grad_norm": 2.5982093811035156,
|
| 660 |
+
"learning_rate": 0.00012938671209540033,
|
| 661 |
+
"loss": 0.6565,
|
| 662 |
+
"step": 830
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.43,
|
| 666 |
+
"grad_norm": 3.1191565990448,
|
| 667 |
+
"learning_rate": 0.00012853492333901194,
|
| 668 |
+
"loss": 0.533,
|
| 669 |
+
"step": 840
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.45,
|
| 673 |
+
"grad_norm": 5.229197025299072,
|
| 674 |
+
"learning_rate": 0.00012768313458262352,
|
| 675 |
+
"loss": 0.6123,
|
| 676 |
+
"step": 850
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.47,
|
| 680 |
+
"grad_norm": 2.259110689163208,
|
| 681 |
+
"learning_rate": 0.0001268313458262351,
|
| 682 |
+
"loss": 0.5183,
|
| 683 |
+
"step": 860
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.48,
|
| 687 |
+
"grad_norm": 3.099496364593506,
|
| 688 |
+
"learning_rate": 0.00012597955706984668,
|
| 689 |
+
"loss": 0.6911,
|
| 690 |
+
"step": 870
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.5,
|
| 694 |
+
"grad_norm": 2.9909987449645996,
|
| 695 |
+
"learning_rate": 0.00012512776831345826,
|
| 696 |
+
"loss": 0.6671,
|
| 697 |
+
"step": 880
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.52,
|
| 701 |
+
"grad_norm": 3.1856462955474854,
|
| 702 |
+
"learning_rate": 0.00012427597955706984,
|
| 703 |
+
"loss": 0.6652,
|
| 704 |
+
"step": 890
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.53,
|
| 708 |
+
"grad_norm": 3.9080755710601807,
|
| 709 |
+
"learning_rate": 0.00012342419080068145,
|
| 710 |
+
"loss": 0.6098,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.53,
|
| 715 |
+
"eval_accuracy": 0.6695059625212947,
|
| 716 |
+
"eval_loss": 0.8459659218788147,
|
| 717 |
+
"eval_runtime": 37.8733,
|
| 718 |
+
"eval_samples_per_second": 30.998,
|
| 719 |
+
"eval_steps_per_second": 3.881,
|
| 720 |
+
"step": 900
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.55,
|
| 724 |
+
"grad_norm": 1.7587580680847168,
|
| 725 |
+
"learning_rate": 0.000122572402044293,
|
| 726 |
+
"loss": 0.7362,
|
| 727 |
+
"step": 910
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.57,
|
| 731 |
+
"grad_norm": 2.7327494621276855,
|
| 732 |
+
"learning_rate": 0.00012172061328790461,
|
| 733 |
+
"loss": 0.5863,
|
| 734 |
+
"step": 920
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.58,
|
| 738 |
+
"grad_norm": 4.113401889801025,
|
| 739 |
+
"learning_rate": 0.0001208688245315162,
|
| 740 |
+
"loss": 0.8205,
|
| 741 |
+
"step": 930
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 1.6,
|
| 745 |
+
"grad_norm": 4.598094940185547,
|
| 746 |
+
"learning_rate": 0.00012001703577512777,
|
| 747 |
+
"loss": 0.7198,
|
| 748 |
+
"step": 940
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 1.62,
|
| 752 |
+
"grad_norm": 2.8792037963867188,
|
| 753 |
+
"learning_rate": 0.00011916524701873937,
|
| 754 |
+
"loss": 0.6532,
|
| 755 |
+
"step": 950
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.64,
|
| 759 |
+
"grad_norm": 2.949414014816284,
|
| 760 |
+
"learning_rate": 0.00011831345826235094,
|
| 761 |
+
"loss": 0.6783,
|
| 762 |
+
"step": 960
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.65,
|
| 766 |
+
"grad_norm": 2.300352096557617,
|
| 767 |
+
"learning_rate": 0.00011746166950596253,
|
| 768 |
+
"loss": 0.69,
|
| 769 |
+
"step": 970
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.67,
|
| 773 |
+
"grad_norm": 2.5100274085998535,
|
| 774 |
+
"learning_rate": 0.00011660988074957411,
|
| 775 |
+
"loss": 0.7028,
|
| 776 |
+
"step": 980
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.69,
|
| 780 |
+
"grad_norm": 2.372359275817871,
|
| 781 |
+
"learning_rate": 0.0001157580919931857,
|
| 782 |
+
"loss": 0.5673,
|
| 783 |
+
"step": 990
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 1.7,
|
| 787 |
+
"grad_norm": 4.268792152404785,
|
| 788 |
+
"learning_rate": 0.00011490630323679727,
|
| 789 |
+
"loss": 0.6251,
|
| 790 |
+
"step": 1000
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.7,
|
| 794 |
+
"eval_accuracy": 0.6703577512776832,
|
| 795 |
+
"eval_loss": 0.8609783053398132,
|
| 796 |
+
"eval_runtime": 37.811,
|
| 797 |
+
"eval_samples_per_second": 31.049,
|
| 798 |
+
"eval_steps_per_second": 3.888,
|
| 799 |
+
"step": 1000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 1.72,
|
| 803 |
+
"grad_norm": 3.081153154373169,
|
| 804 |
+
"learning_rate": 0.00011405451448040887,
|
| 805 |
+
"loss": 0.7021,
|
| 806 |
+
"step": 1010
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 1.74,
|
| 810 |
+
"grad_norm": 2.9631364345550537,
|
| 811 |
+
"learning_rate": 0.0001132879045996593,
|
| 812 |
+
"loss": 0.5469,
|
| 813 |
+
"step": 1020
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 1.75,
|
| 817 |
+
"grad_norm": 3.2896649837493896,
|
| 818 |
+
"learning_rate": 0.00011243611584327087,
|
| 819 |
+
"loss": 0.5593,
|
| 820 |
+
"step": 1030
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 1.77,
|
| 824 |
+
"grad_norm": 3.8375134468078613,
|
| 825 |
+
"learning_rate": 0.00011158432708688246,
|
| 826 |
+
"loss": 0.5499,
|
| 827 |
+
"step": 1040
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.79,
|
| 831 |
+
"grad_norm": 1.5597748756408691,
|
| 832 |
+
"learning_rate": 0.00011073253833049404,
|
| 833 |
+
"loss": 0.5529,
|
| 834 |
+
"step": 1050
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.81,
|
| 838 |
+
"grad_norm": 4.54299783706665,
|
| 839 |
+
"learning_rate": 0.00010988074957410564,
|
| 840 |
+
"loss": 0.6211,
|
| 841 |
+
"step": 1060
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.82,
|
| 845 |
+
"grad_norm": 3.2734501361846924,
|
| 846 |
+
"learning_rate": 0.0001090289608177172,
|
| 847 |
+
"loss": 0.7002,
|
| 848 |
+
"step": 1070
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 1.84,
|
| 852 |
+
"grad_norm": 3.7582859992980957,
|
| 853 |
+
"learning_rate": 0.0001081771720613288,
|
| 854 |
+
"loss": 0.7465,
|
| 855 |
+
"step": 1080
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 1.86,
|
| 859 |
+
"grad_norm": 2.190544605255127,
|
| 860 |
+
"learning_rate": 0.00010732538330494038,
|
| 861 |
+
"loss": 0.6662,
|
| 862 |
+
"step": 1090
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 1.87,
|
| 866 |
+
"grad_norm": 1.7477951049804688,
|
| 867 |
+
"learning_rate": 0.00010647359454855197,
|
| 868 |
+
"loss": 0.7863,
|
| 869 |
+
"step": 1100
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 1.87,
|
| 873 |
+
"eval_accuracy": 0.6431005110732538,
|
| 874 |
+
"eval_loss": 0.8668282628059387,
|
| 875 |
+
"eval_runtime": 37.5178,
|
| 876 |
+
"eval_samples_per_second": 31.292,
|
| 877 |
+
"eval_steps_per_second": 3.918,
|
| 878 |
+
"step": 1100
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.89,
|
| 882 |
+
"grad_norm": 1.9970145225524902,
|
| 883 |
+
"learning_rate": 0.00010562180579216354,
|
| 884 |
+
"loss": 0.5988,
|
| 885 |
+
"step": 1110
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.91,
|
| 889 |
+
"grad_norm": 3.718055248260498,
|
| 890 |
+
"learning_rate": 0.00010477001703577514,
|
| 891 |
+
"loss": 0.5973,
|
| 892 |
+
"step": 1120
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.93,
|
| 896 |
+
"grad_norm": 1.6347967386245728,
|
| 897 |
+
"learning_rate": 0.0001039182282793867,
|
| 898 |
+
"loss": 0.5818,
|
| 899 |
+
"step": 1130
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.94,
|
| 903 |
+
"grad_norm": 2.3118577003479004,
|
| 904 |
+
"learning_rate": 0.0001030664395229983,
|
| 905 |
+
"loss": 0.5136,
|
| 906 |
+
"step": 1140
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.96,
|
| 910 |
+
"grad_norm": 2.806833267211914,
|
| 911 |
+
"learning_rate": 0.00010221465076660988,
|
| 912 |
+
"loss": 0.5353,
|
| 913 |
+
"step": 1150
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.98,
|
| 917 |
+
"grad_norm": 2.699890375137329,
|
| 918 |
+
"learning_rate": 0.00010136286201022147,
|
| 919 |
+
"loss": 0.5498,
|
| 920 |
+
"step": 1160
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.99,
|
| 924 |
+
"grad_norm": 2.3461856842041016,
|
| 925 |
+
"learning_rate": 0.00010051107325383304,
|
| 926 |
+
"loss": 0.7181,
|
| 927 |
+
"step": 1170
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.01,
|
| 931 |
+
"grad_norm": 2.483959436416626,
|
| 932 |
+
"learning_rate": 9.965928449744463e-05,
|
| 933 |
+
"loss": 0.3872,
|
| 934 |
+
"step": 1180
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 2.03,
|
| 938 |
+
"grad_norm": 2.1393377780914307,
|
| 939 |
+
"learning_rate": 9.880749574105622e-05,
|
| 940 |
+
"loss": 0.292,
|
| 941 |
+
"step": 1190
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 2.04,
|
| 945 |
+
"grad_norm": 1.6828927993774414,
|
| 946 |
+
"learning_rate": 9.795570698466781e-05,
|
| 947 |
+
"loss": 0.2595,
|
| 948 |
+
"step": 1200
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 2.04,
|
| 952 |
+
"eval_accuracy": 0.6839863713798978,
|
| 953 |
+
"eval_loss": 0.8725138902664185,
|
| 954 |
+
"eval_runtime": 37.2408,
|
| 955 |
+
"eval_samples_per_second": 31.525,
|
| 956 |
+
"eval_steps_per_second": 3.947,
|
| 957 |
+
"step": 1200
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 2.06,
|
| 961 |
+
"grad_norm": 1.973240852355957,
|
| 962 |
+
"learning_rate": 9.710391822827939e-05,
|
| 963 |
+
"loss": 0.2644,
|
| 964 |
+
"step": 1210
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 2.08,
|
| 968 |
+
"grad_norm": 2.932751417160034,
|
| 969 |
+
"learning_rate": 9.625212947189097e-05,
|
| 970 |
+
"loss": 0.2925,
|
| 971 |
+
"step": 1220
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.1,
|
| 975 |
+
"grad_norm": 3.356760025024414,
|
| 976 |
+
"learning_rate": 9.540034071550255e-05,
|
| 977 |
+
"loss": 0.312,
|
| 978 |
+
"step": 1230
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 2.11,
|
| 982 |
+
"grad_norm": 1.7125446796417236,
|
| 983 |
+
"learning_rate": 9.454855195911415e-05,
|
| 984 |
+
"loss": 0.2277,
|
| 985 |
+
"step": 1240
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 2.13,
|
| 989 |
+
"grad_norm": 1.714805006980896,
|
| 990 |
+
"learning_rate": 9.369676320272573e-05,
|
| 991 |
+
"loss": 0.3301,
|
| 992 |
+
"step": 1250
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.15,
|
| 996 |
+
"grad_norm": 2.301734685897827,
|
| 997 |
+
"learning_rate": 9.284497444633732e-05,
|
| 998 |
+
"loss": 0.2668,
|
| 999 |
+
"step": 1260
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.16,
|
| 1003 |
+
"grad_norm": 2.4843878746032715,
|
| 1004 |
+
"learning_rate": 9.19931856899489e-05,
|
| 1005 |
+
"loss": 0.2333,
|
| 1006 |
+
"step": 1270
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 2.18,
|
| 1010 |
+
"grad_norm": 2.9054977893829346,
|
| 1011 |
+
"learning_rate": 9.114139693356048e-05,
|
| 1012 |
+
"loss": 0.3492,
|
| 1013 |
+
"step": 1280
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 2.2,
|
| 1017 |
+
"grad_norm": 4.664933681488037,
|
| 1018 |
+
"learning_rate": 9.028960817717206e-05,
|
| 1019 |
+
"loss": 0.3754,
|
| 1020 |
+
"step": 1290
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 2.21,
|
| 1024 |
+
"grad_norm": 2.1164679527282715,
|
| 1025 |
+
"learning_rate": 8.943781942078366e-05,
|
| 1026 |
+
"loss": 0.2735,
|
| 1027 |
+
"step": 1300
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 2.21,
|
| 1031 |
+
"eval_accuracy": 0.6746166950596252,
|
| 1032 |
+
"eval_loss": 0.9306557178497314,
|
| 1033 |
+
"eval_runtime": 37.0944,
|
| 1034 |
+
"eval_samples_per_second": 31.649,
|
| 1035 |
+
"eval_steps_per_second": 3.963,
|
| 1036 |
+
"step": 1300
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 2.23,
|
| 1040 |
+
"grad_norm": 4.541740894317627,
|
| 1041 |
+
"learning_rate": 8.858603066439524e-05,
|
| 1042 |
+
"loss": 0.3835,
|
| 1043 |
+
"step": 1310
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 2.25,
|
| 1047 |
+
"grad_norm": 3.0828359127044678,
|
| 1048 |
+
"learning_rate": 8.773424190800682e-05,
|
| 1049 |
+
"loss": 0.3189,
|
| 1050 |
+
"step": 1320
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 2.27,
|
| 1054 |
+
"grad_norm": 2.398512363433838,
|
| 1055 |
+
"learning_rate": 8.68824531516184e-05,
|
| 1056 |
+
"loss": 0.29,
|
| 1057 |
+
"step": 1330
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 2.28,
|
| 1061 |
+
"grad_norm": 3.069840908050537,
|
| 1062 |
+
"learning_rate": 8.603066439522998e-05,
|
| 1063 |
+
"loss": 0.288,
|
| 1064 |
+
"step": 1340
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 2.3,
|
| 1068 |
+
"grad_norm": 5.078506946563721,
|
| 1069 |
+
"learning_rate": 8.517887563884158e-05,
|
| 1070 |
+
"loss": 0.2772,
|
| 1071 |
+
"step": 1350
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 2.32,
|
| 1075 |
+
"grad_norm": 2.812199354171753,
|
| 1076 |
+
"learning_rate": 8.432708688245316e-05,
|
| 1077 |
+
"loss": 0.2951,
|
| 1078 |
+
"step": 1360
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 2.33,
|
| 1082 |
+
"grad_norm": 4.542017936706543,
|
| 1083 |
+
"learning_rate": 8.347529812606474e-05,
|
| 1084 |
+
"loss": 0.2142,
|
| 1085 |
+
"step": 1370
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 2.35,
|
| 1089 |
+
"grad_norm": 3.7486696243286133,
|
| 1090 |
+
"learning_rate": 8.262350936967632e-05,
|
| 1091 |
+
"loss": 0.257,
|
| 1092 |
+
"step": 1380
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 2.37,
|
| 1096 |
+
"grad_norm": 3.5566983222961426,
|
| 1097 |
+
"learning_rate": 8.17717206132879e-05,
|
| 1098 |
+
"loss": 0.2816,
|
| 1099 |
+
"step": 1390
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 2.39,
|
| 1103 |
+
"grad_norm": 1.3465384244918823,
|
| 1104 |
+
"learning_rate": 8.09199318568995e-05,
|
| 1105 |
+
"loss": 0.2429,
|
| 1106 |
+
"step": 1400
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 2.39,
|
| 1110 |
+
"eval_accuracy": 0.6354344122657581,
|
| 1111 |
+
"eval_loss": 1.0957823991775513,
|
| 1112 |
+
"eval_runtime": 37.2033,
|
| 1113 |
+
"eval_samples_per_second": 31.556,
|
| 1114 |
+
"eval_steps_per_second": 3.951,
|
| 1115 |
+
"step": 1400
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.4,
|
| 1119 |
+
"grad_norm": 2.310131788253784,
|
| 1120 |
+
"learning_rate": 8.006814310051108e-05,
|
| 1121 |
+
"loss": 0.306,
|
| 1122 |
+
"step": 1410
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.42,
|
| 1126 |
+
"grad_norm": 3.1297261714935303,
|
| 1127 |
+
"learning_rate": 7.921635434412266e-05,
|
| 1128 |
+
"loss": 0.3257,
|
| 1129 |
+
"step": 1420
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.44,
|
| 1133 |
+
"grad_norm": 1.8082480430603027,
|
| 1134 |
+
"learning_rate": 7.836456558773425e-05,
|
| 1135 |
+
"loss": 0.2001,
|
| 1136 |
+
"step": 1430
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.45,
|
| 1140 |
+
"grad_norm": 1.7700148820877075,
|
| 1141 |
+
"learning_rate": 7.751277683134583e-05,
|
| 1142 |
+
"loss": 0.3476,
|
| 1143 |
+
"step": 1440
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.47,
|
| 1147 |
+
"grad_norm": 4.247625350952148,
|
| 1148 |
+
"learning_rate": 7.666098807495741e-05,
|
| 1149 |
+
"loss": 0.2323,
|
| 1150 |
+
"step": 1450
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.49,
|
| 1154 |
+
"grad_norm": 4.059571743011475,
|
| 1155 |
+
"learning_rate": 7.5809199318569e-05,
|
| 1156 |
+
"loss": 0.3089,
|
| 1157 |
+
"step": 1460
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.5,
|
| 1161 |
+
"grad_norm": 3.2417612075805664,
|
| 1162 |
+
"learning_rate": 7.495741056218059e-05,
|
| 1163 |
+
"loss": 0.1964,
|
| 1164 |
+
"step": 1470
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.52,
|
| 1168 |
+
"grad_norm": 5.7817463874816895,
|
| 1169 |
+
"learning_rate": 7.410562180579217e-05,
|
| 1170 |
+
"loss": 0.3549,
|
| 1171 |
+
"step": 1480
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.54,
|
| 1175 |
+
"grad_norm": 5.440825939178467,
|
| 1176 |
+
"learning_rate": 7.325383304940375e-05,
|
| 1177 |
+
"loss": 0.3085,
|
| 1178 |
+
"step": 1490
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.56,
|
| 1182 |
+
"grad_norm": 4.482067108154297,
|
| 1183 |
+
"learning_rate": 7.240204429301533e-05,
|
| 1184 |
+
"loss": 0.3224,
|
| 1185 |
+
"step": 1500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.56,
|
| 1189 |
+
"eval_accuracy": 0.6686541737649063,
|
| 1190 |
+
"eval_loss": 1.0305246114730835,
|
| 1191 |
+
"eval_runtime": 37.1181,
|
| 1192 |
+
"eval_samples_per_second": 31.629,
|
| 1193 |
+
"eval_steps_per_second": 3.96,
|
| 1194 |
+
"step": 1500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.57,
|
| 1198 |
+
"grad_norm": 2.1568057537078857,
|
| 1199 |
+
"learning_rate": 7.155025553662692e-05,
|
| 1200 |
+
"loss": 0.1612,
|
| 1201 |
+
"step": 1510
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 2.59,
|
| 1205 |
+
"grad_norm": 1.293427586555481,
|
| 1206 |
+
"learning_rate": 7.06984667802385e-05,
|
| 1207 |
+
"loss": 0.3217,
|
| 1208 |
+
"step": 1520
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 2.61,
|
| 1212 |
+
"grad_norm": 4.301244258880615,
|
| 1213 |
+
"learning_rate": 6.984667802385009e-05,
|
| 1214 |
+
"loss": 0.2378,
|
| 1215 |
+
"step": 1530
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 2.62,
|
| 1219 |
+
"grad_norm": 1.6040468215942383,
|
| 1220 |
+
"learning_rate": 6.899488926746167e-05,
|
| 1221 |
+
"loss": 0.2801,
|
| 1222 |
+
"step": 1540
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 2.64,
|
| 1226 |
+
"grad_norm": 0.7993047833442688,
|
| 1227 |
+
"learning_rate": 6.814310051107326e-05,
|
| 1228 |
+
"loss": 0.2637,
|
| 1229 |
+
"step": 1550
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 2.66,
|
| 1233 |
+
"grad_norm": 4.865533828735352,
|
| 1234 |
+
"learning_rate": 6.729131175468484e-05,
|
| 1235 |
+
"loss": 0.3441,
|
| 1236 |
+
"step": 1560
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 2.67,
|
| 1240 |
+
"grad_norm": 1.7501546144485474,
|
| 1241 |
+
"learning_rate": 6.643952299829642e-05,
|
| 1242 |
+
"loss": 0.2523,
|
| 1243 |
+
"step": 1570
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 2.69,
|
| 1247 |
+
"grad_norm": 1.331475019454956,
|
| 1248 |
+
"learning_rate": 6.5587734241908e-05,
|
| 1249 |
+
"loss": 0.2127,
|
| 1250 |
+
"step": 1580
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 2.71,
|
| 1254 |
+
"grad_norm": 3.352147102355957,
|
| 1255 |
+
"learning_rate": 6.473594548551958e-05,
|
| 1256 |
+
"loss": 0.3432,
|
| 1257 |
+
"step": 1590
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 2.73,
|
| 1261 |
+
"grad_norm": 0.3470512330532074,
|
| 1262 |
+
"learning_rate": 6.388415672913118e-05,
|
| 1263 |
+
"loss": 0.1602,
|
| 1264 |
+
"step": 1600
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 2.73,
|
| 1268 |
+
"eval_accuracy": 0.6746166950596252,
|
| 1269 |
+
"eval_loss": 1.0072139501571655,
|
| 1270 |
+
"eval_runtime": 37.0019,
|
| 1271 |
+
"eval_samples_per_second": 31.728,
|
| 1272 |
+
"eval_steps_per_second": 3.973,
|
| 1273 |
+
"step": 1600
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 2.74,
|
| 1277 |
+
"grad_norm": 3.1594250202178955,
|
| 1278 |
+
"learning_rate": 6.303236797274277e-05,
|
| 1279 |
+
"loss": 0.1929,
|
| 1280 |
+
"step": 1610
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 2.76,
|
| 1284 |
+
"grad_norm": 4.477923393249512,
|
| 1285 |
+
"learning_rate": 6.218057921635435e-05,
|
| 1286 |
+
"loss": 0.2696,
|
| 1287 |
+
"step": 1620
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 2.78,
|
| 1291 |
+
"grad_norm": 3.042938232421875,
|
| 1292 |
+
"learning_rate": 6.132879045996594e-05,
|
| 1293 |
+
"loss": 0.2527,
|
| 1294 |
+
"step": 1630
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 2.79,
|
| 1298 |
+
"grad_norm": 0.8534514904022217,
|
| 1299 |
+
"learning_rate": 6.0477001703577516e-05,
|
| 1300 |
+
"loss": 0.1727,
|
| 1301 |
+
"step": 1640
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 2.81,
|
| 1305 |
+
"grad_norm": 2.2307116985321045,
|
| 1306 |
+
"learning_rate": 5.9625212947189104e-05,
|
| 1307 |
+
"loss": 0.3178,
|
| 1308 |
+
"step": 1650
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 2.83,
|
| 1312 |
+
"grad_norm": 3.302003860473633,
|
| 1313 |
+
"learning_rate": 5.8773424190800684e-05,
|
| 1314 |
+
"loss": 0.2973,
|
| 1315 |
+
"step": 1660
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 2.84,
|
| 1319 |
+
"grad_norm": 5.320656776428223,
|
| 1320 |
+
"learning_rate": 5.792163543441227e-05,
|
| 1321 |
+
"loss": 0.2715,
|
| 1322 |
+
"step": 1670
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 2.86,
|
| 1326 |
+
"grad_norm": 3.923163414001465,
|
| 1327 |
+
"learning_rate": 5.706984667802385e-05,
|
| 1328 |
+
"loss": 0.1991,
|
| 1329 |
+
"step": 1680
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.88,
|
| 1333 |
+
"grad_norm": 7.479254245758057,
|
| 1334 |
+
"learning_rate": 5.6218057921635434e-05,
|
| 1335 |
+
"loss": 0.321,
|
| 1336 |
+
"step": 1690
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 2.9,
|
| 1340 |
+
"grad_norm": 2.2710225582122803,
|
| 1341 |
+
"learning_rate": 5.536626916524702e-05,
|
| 1342 |
+
"loss": 0.2042,
|
| 1343 |
+
"step": 1700
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 2.9,
|
| 1347 |
+
"eval_accuracy": 0.6788756388415673,
|
| 1348 |
+
"eval_loss": 1.0971218347549438,
|
| 1349 |
+
"eval_runtime": 36.9173,
|
| 1350 |
+
"eval_samples_per_second": 31.801,
|
| 1351 |
+
"eval_steps_per_second": 3.982,
|
| 1352 |
+
"step": 1700
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 2.91,
|
| 1356 |
+
"grad_norm": 2.7610058784484863,
|
| 1357 |
+
"learning_rate": 5.45144804088586e-05,
|
| 1358 |
+
"loss": 0.3396,
|
| 1359 |
+
"step": 1710
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 2.93,
|
| 1363 |
+
"grad_norm": 2.2475104331970215,
|
| 1364 |
+
"learning_rate": 5.366269165247019e-05,
|
| 1365 |
+
"loss": 0.266,
|
| 1366 |
+
"step": 1720
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 2.95,
|
| 1370 |
+
"grad_norm": 4.55673885345459,
|
| 1371 |
+
"learning_rate": 5.281090289608177e-05,
|
| 1372 |
+
"loss": 0.341,
|
| 1373 |
+
"step": 1730
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 2.96,
|
| 1377 |
+
"grad_norm": 4.0248260498046875,
|
| 1378 |
+
"learning_rate": 5.195911413969335e-05,
|
| 1379 |
+
"loss": 0.2005,
|
| 1380 |
+
"step": 1740
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 2.98,
|
| 1384 |
+
"grad_norm": 4.798257827758789,
|
| 1385 |
+
"learning_rate": 5.110732538330494e-05,
|
| 1386 |
+
"loss": 0.2615,
|
| 1387 |
+
"step": 1750
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 3.0,
|
| 1391 |
+
"grad_norm": 3.2967402935028076,
|
| 1392 |
+
"learning_rate": 5.025553662691652e-05,
|
| 1393 |
+
"loss": 0.1966,
|
| 1394 |
+
"step": 1760
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 3.02,
|
| 1398 |
+
"grad_norm": 5.774517059326172,
|
| 1399 |
+
"learning_rate": 4.940374787052811e-05,
|
| 1400 |
+
"loss": 0.1141,
|
| 1401 |
+
"step": 1770
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 3.03,
|
| 1405 |
+
"grad_norm": 1.7739803791046143,
|
| 1406 |
+
"learning_rate": 4.8551959114139695e-05,
|
| 1407 |
+
"loss": 0.0671,
|
| 1408 |
+
"step": 1780
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 3.05,
|
| 1412 |
+
"grad_norm": 0.8837150931358337,
|
| 1413 |
+
"learning_rate": 4.7700170357751276e-05,
|
| 1414 |
+
"loss": 0.0835,
|
| 1415 |
+
"step": 1790
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 3.07,
|
| 1419 |
+
"grad_norm": 1.7833037376403809,
|
| 1420 |
+
"learning_rate": 4.6848381601362864e-05,
|
| 1421 |
+
"loss": 0.0604,
|
| 1422 |
+
"step": 1800
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 3.07,
|
| 1426 |
+
"eval_accuracy": 0.6916524701873935,
|
| 1427 |
+
"eval_loss": 1.0816737413406372,
|
| 1428 |
+
"eval_runtime": 36.8222,
|
| 1429 |
+
"eval_samples_per_second": 31.883,
|
| 1430 |
+
"eval_steps_per_second": 3.992,
|
| 1431 |
+
"step": 1800
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 3.08,
|
| 1435 |
+
"grad_norm": 0.34585830569267273,
|
| 1436 |
+
"learning_rate": 4.599659284497445e-05,
|
| 1437 |
+
"loss": 0.092,
|
| 1438 |
+
"step": 1810
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 3.1,
|
| 1442 |
+
"grad_norm": 0.7962571382522583,
|
| 1443 |
+
"learning_rate": 4.514480408858603e-05,
|
| 1444 |
+
"loss": 0.0587,
|
| 1445 |
+
"step": 1820
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 3.12,
|
| 1449 |
+
"grad_norm": 0.16402888298034668,
|
| 1450 |
+
"learning_rate": 4.429301533219762e-05,
|
| 1451 |
+
"loss": 0.0547,
|
| 1452 |
+
"step": 1830
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 3.13,
|
| 1456 |
+
"grad_norm": 0.624047040939331,
|
| 1457 |
+
"learning_rate": 4.34412265758092e-05,
|
| 1458 |
+
"loss": 0.0954,
|
| 1459 |
+
"step": 1840
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 3.15,
|
| 1463 |
+
"grad_norm": 0.4253842532634735,
|
| 1464 |
+
"learning_rate": 4.258943781942079e-05,
|
| 1465 |
+
"loss": 0.0567,
|
| 1466 |
+
"step": 1850
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 3.17,
|
| 1470 |
+
"grad_norm": 0.1523701399564743,
|
| 1471 |
+
"learning_rate": 4.173764906303237e-05,
|
| 1472 |
+
"loss": 0.0413,
|
| 1473 |
+
"step": 1860
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 3.19,
|
| 1477 |
+
"grad_norm": 4.592818260192871,
|
| 1478 |
+
"learning_rate": 4.088586030664395e-05,
|
| 1479 |
+
"loss": 0.0968,
|
| 1480 |
+
"step": 1870
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 3.2,
|
| 1484 |
+
"grad_norm": 1.4066344499588013,
|
| 1485 |
+
"learning_rate": 4.003407155025554e-05,
|
| 1486 |
+
"loss": 0.1454,
|
| 1487 |
+
"step": 1880
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 3.22,
|
| 1491 |
+
"grad_norm": 2.1996095180511475,
|
| 1492 |
+
"learning_rate": 3.9182282793867125e-05,
|
| 1493 |
+
"loss": 0.1128,
|
| 1494 |
+
"step": 1890
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 3.24,
|
| 1498 |
+
"grad_norm": 0.102027028799057,
|
| 1499 |
+
"learning_rate": 3.8330494037478706e-05,
|
| 1500 |
+
"loss": 0.0716,
|
| 1501 |
+
"step": 1900
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 3.24,
|
| 1505 |
+
"eval_accuracy": 0.692504258943782,
|
| 1506 |
+
"eval_loss": 1.1307132244110107,
|
| 1507 |
+
"eval_runtime": 37.0403,
|
| 1508 |
+
"eval_samples_per_second": 31.695,
|
| 1509 |
+
"eval_steps_per_second": 3.969,
|
| 1510 |
+
"step": 1900
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 3.25,
|
| 1514 |
+
"grad_norm": 1.6857343912124634,
|
| 1515 |
+
"learning_rate": 3.7478705281090294e-05,
|
| 1516 |
+
"loss": 0.04,
|
| 1517 |
+
"step": 1910
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 3.27,
|
| 1521 |
+
"grad_norm": 1.2973403930664062,
|
| 1522 |
+
"learning_rate": 3.6626916524701875e-05,
|
| 1523 |
+
"loss": 0.0403,
|
| 1524 |
+
"step": 1920
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 3.29,
|
| 1528 |
+
"grad_norm": 0.41860514879226685,
|
| 1529 |
+
"learning_rate": 3.577512776831346e-05,
|
| 1530 |
+
"loss": 0.0642,
|
| 1531 |
+
"step": 1930
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 3.3,
|
| 1535 |
+
"grad_norm": 0.5436795353889465,
|
| 1536 |
+
"learning_rate": 3.492333901192504e-05,
|
| 1537 |
+
"loss": 0.0836,
|
| 1538 |
+
"step": 1940
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 3.32,
|
| 1542 |
+
"grad_norm": 0.21996204555034637,
|
| 1543 |
+
"learning_rate": 3.407155025553663e-05,
|
| 1544 |
+
"loss": 0.0406,
|
| 1545 |
+
"step": 1950
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 3.34,
|
| 1549 |
+
"grad_norm": 0.14845231175422668,
|
| 1550 |
+
"learning_rate": 3.321976149914821e-05,
|
| 1551 |
+
"loss": 0.0385,
|
| 1552 |
+
"step": 1960
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 3.36,
|
| 1556 |
+
"grad_norm": 3.531405448913574,
|
| 1557 |
+
"learning_rate": 3.236797274275979e-05,
|
| 1558 |
+
"loss": 0.0824,
|
| 1559 |
+
"step": 1970
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 3.37,
|
| 1563 |
+
"grad_norm": 0.07682117819786072,
|
| 1564 |
+
"learning_rate": 3.151618398637139e-05,
|
| 1565 |
+
"loss": 0.0717,
|
| 1566 |
+
"step": 1980
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 3.39,
|
| 1570 |
+
"grad_norm": 0.07611515372991562,
|
| 1571 |
+
"learning_rate": 3.066439522998297e-05,
|
| 1572 |
+
"loss": 0.0572,
|
| 1573 |
+
"step": 1990
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 3.41,
|
| 1577 |
+
"grad_norm": 0.6266534328460693,
|
| 1578 |
+
"learning_rate": 2.9812606473594552e-05,
|
| 1579 |
+
"loss": 0.0822,
|
| 1580 |
+
"step": 2000
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 3.41,
|
| 1584 |
+
"eval_accuracy": 0.692504258943782,
|
| 1585 |
+
"eval_loss": 1.1826940774917603,
|
| 1586 |
+
"eval_runtime": 37.1369,
|
| 1587 |
+
"eval_samples_per_second": 31.613,
|
| 1588 |
+
"eval_steps_per_second": 3.958,
|
| 1589 |
+
"step": 2000
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 3.42,
|
| 1593 |
+
"grad_norm": 0.1280030608177185,
|
| 1594 |
+
"learning_rate": 2.8960817717206136e-05,
|
| 1595 |
+
"loss": 0.0244,
|
| 1596 |
+
"step": 2010
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 3.44,
|
| 1600 |
+
"grad_norm": 0.07406999170780182,
|
| 1601 |
+
"learning_rate": 2.8109028960817717e-05,
|
| 1602 |
+
"loss": 0.0574,
|
| 1603 |
+
"step": 2020
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 3.46,
|
| 1607 |
+
"grad_norm": 5.587332248687744,
|
| 1608 |
+
"learning_rate": 2.72572402044293e-05,
|
| 1609 |
+
"loss": 0.0352,
|
| 1610 |
+
"step": 2030
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 3.48,
|
| 1614 |
+
"grad_norm": 2.2010979652404785,
|
| 1615 |
+
"learning_rate": 2.6405451448040885e-05,
|
| 1616 |
+
"loss": 0.0789,
|
| 1617 |
+
"step": 2040
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 3.49,
|
| 1621 |
+
"grad_norm": 2.9271368980407715,
|
| 1622 |
+
"learning_rate": 2.555366269165247e-05,
|
| 1623 |
+
"loss": 0.082,
|
| 1624 |
+
"step": 2050
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 3.51,
|
| 1628 |
+
"grad_norm": 0.05890679359436035,
|
| 1629 |
+
"learning_rate": 2.4701873935264054e-05,
|
| 1630 |
+
"loss": 0.0769,
|
| 1631 |
+
"step": 2060
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 3.53,
|
| 1635 |
+
"grad_norm": 0.7043523192405701,
|
| 1636 |
+
"learning_rate": 2.3850085178875638e-05,
|
| 1637 |
+
"loss": 0.0819,
|
| 1638 |
+
"step": 2070
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 3.54,
|
| 1642 |
+
"grad_norm": 0.12047506123781204,
|
| 1643 |
+
"learning_rate": 2.2998296422487226e-05,
|
| 1644 |
+
"loss": 0.0195,
|
| 1645 |
+
"step": 2080
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 3.56,
|
| 1649 |
+
"grad_norm": 0.1116802990436554,
|
| 1650 |
+
"learning_rate": 2.214650766609881e-05,
|
| 1651 |
+
"loss": 0.0159,
|
| 1652 |
+
"step": 2090
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 3.58,
|
| 1656 |
+
"grad_norm": 0.09187493473291397,
|
| 1657 |
+
"learning_rate": 2.1294718909710394e-05,
|
| 1658 |
+
"loss": 0.0889,
|
| 1659 |
+
"step": 2100
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 3.58,
|
| 1663 |
+
"eval_accuracy": 0.6933560477001703,
|
| 1664 |
+
"eval_loss": 1.2423571348190308,
|
| 1665 |
+
"eval_runtime": 37.3059,
|
| 1666 |
+
"eval_samples_per_second": 31.47,
|
| 1667 |
+
"eval_steps_per_second": 3.94,
|
| 1668 |
+
"step": 2100
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 3.59,
|
| 1672 |
+
"grad_norm": 4.332376956939697,
|
| 1673 |
+
"learning_rate": 2.0442930153321975e-05,
|
| 1674 |
+
"loss": 0.0939,
|
| 1675 |
+
"step": 2110
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 3.61,
|
| 1679 |
+
"grad_norm": 0.13916102051734924,
|
| 1680 |
+
"learning_rate": 1.9591141396933563e-05,
|
| 1681 |
+
"loss": 0.0933,
|
| 1682 |
+
"step": 2120
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 3.63,
|
| 1686 |
+
"grad_norm": 7.690703392028809,
|
| 1687 |
+
"learning_rate": 1.8739352640545147e-05,
|
| 1688 |
+
"loss": 0.0496,
|
| 1689 |
+
"step": 2130
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 3.65,
|
| 1693 |
+
"grad_norm": 2.5700595378875732,
|
| 1694 |
+
"learning_rate": 1.788756388415673e-05,
|
| 1695 |
+
"loss": 0.0782,
|
| 1696 |
+
"step": 2140
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 3.66,
|
| 1700 |
+
"grad_norm": 0.20934216678142548,
|
| 1701 |
+
"learning_rate": 1.7035775127768315e-05,
|
| 1702 |
+
"loss": 0.0606,
|
| 1703 |
+
"step": 2150
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 3.68,
|
| 1707 |
+
"grad_norm": 1.2959486246109009,
|
| 1708 |
+
"learning_rate": 1.6183986371379896e-05,
|
| 1709 |
+
"loss": 0.0601,
|
| 1710 |
+
"step": 2160
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 3.7,
|
| 1714 |
+
"grad_norm": 0.2652721405029297,
|
| 1715 |
+
"learning_rate": 1.5332197614991484e-05,
|
| 1716 |
+
"loss": 0.062,
|
| 1717 |
+
"step": 2170
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 3.71,
|
| 1721 |
+
"grad_norm": 0.48360127210617065,
|
| 1722 |
+
"learning_rate": 1.4480408858603068e-05,
|
| 1723 |
+
"loss": 0.054,
|
| 1724 |
+
"step": 2180
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 3.73,
|
| 1728 |
+
"grad_norm": 3.1118693351745605,
|
| 1729 |
+
"learning_rate": 1.362862010221465e-05,
|
| 1730 |
+
"loss": 0.0989,
|
| 1731 |
+
"step": 2190
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 3.75,
|
| 1735 |
+
"grad_norm": 0.9077383279800415,
|
| 1736 |
+
"learning_rate": 1.2776831345826235e-05,
|
| 1737 |
+
"loss": 0.0855,
|
| 1738 |
+
"step": 2200
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 3.75,
|
| 1742 |
+
"eval_accuracy": 0.6899488926746167,
|
| 1743 |
+
"eval_loss": 1.2667156457901,
|
| 1744 |
+
"eval_runtime": 36.8511,
|
| 1745 |
+
"eval_samples_per_second": 31.858,
|
| 1746 |
+
"eval_steps_per_second": 3.989,
|
| 1747 |
+
"step": 2200
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 3.76,
|
| 1751 |
+
"grad_norm": 0.13304296135902405,
|
| 1752 |
+
"learning_rate": 1.1925042589437819e-05,
|
| 1753 |
+
"loss": 0.0675,
|
| 1754 |
+
"step": 2210
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 3.78,
|
| 1758 |
+
"grad_norm": 1.3241567611694336,
|
| 1759 |
+
"learning_rate": 1.1073253833049405e-05,
|
| 1760 |
+
"loss": 0.0753,
|
| 1761 |
+
"step": 2220
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 3.8,
|
| 1765 |
+
"grad_norm": 0.2818525731563568,
|
| 1766 |
+
"learning_rate": 1.0221465076660987e-05,
|
| 1767 |
+
"loss": 0.0998,
|
| 1768 |
+
"step": 2230
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 3.82,
|
| 1772 |
+
"grad_norm": 7.136697292327881,
|
| 1773 |
+
"learning_rate": 9.369676320272573e-06,
|
| 1774 |
+
"loss": 0.0314,
|
| 1775 |
+
"step": 2240
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 3.83,
|
| 1779 |
+
"grad_norm": 1.372044324874878,
|
| 1780 |
+
"learning_rate": 8.517887563884158e-06,
|
| 1781 |
+
"loss": 0.0768,
|
| 1782 |
+
"step": 2250
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 3.85,
|
| 1786 |
+
"grad_norm": 6.264348983764648,
|
| 1787 |
+
"learning_rate": 7.666098807495742e-06,
|
| 1788 |
+
"loss": 0.1516,
|
| 1789 |
+
"step": 2260
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 3.87,
|
| 1793 |
+
"grad_norm": 0.1342085599899292,
|
| 1794 |
+
"learning_rate": 6.814310051107325e-06,
|
| 1795 |
+
"loss": 0.0812,
|
| 1796 |
+
"step": 2270
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 3.88,
|
| 1800 |
+
"grad_norm": 0.7664629220962524,
|
| 1801 |
+
"learning_rate": 5.9625212947189095e-06,
|
| 1802 |
+
"loss": 0.0474,
|
| 1803 |
+
"step": 2280
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 3.9,
|
| 1807 |
+
"grad_norm": 4.264090538024902,
|
| 1808 |
+
"learning_rate": 5.110732538330494e-06,
|
| 1809 |
+
"loss": 0.0903,
|
| 1810 |
+
"step": 2290
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 3.92,
|
| 1814 |
+
"grad_norm": 0.07316776365041733,
|
| 1815 |
+
"learning_rate": 4.258943781942079e-06,
|
| 1816 |
+
"loss": 0.0682,
|
| 1817 |
+
"step": 2300
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 3.92,
|
| 1821 |
+
"eval_accuracy": 0.6950596252129472,
|
| 1822 |
+
"eval_loss": 1.2470241785049438,
|
| 1823 |
+
"eval_runtime": 37.0027,
|
| 1824 |
+
"eval_samples_per_second": 31.727,
|
| 1825 |
+
"eval_steps_per_second": 3.973,
|
| 1826 |
+
"step": 2300
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 3.94,
|
| 1830 |
+
"grad_norm": 1.477973222732544,
|
| 1831 |
+
"learning_rate": 3.4071550255536626e-06,
|
| 1832 |
+
"loss": 0.0587,
|
| 1833 |
+
"step": 2310
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 3.95,
|
| 1837 |
+
"grad_norm": 1.249779224395752,
|
| 1838 |
+
"learning_rate": 2.555366269165247e-06,
|
| 1839 |
+
"loss": 0.0546,
|
| 1840 |
+
"step": 2320
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 3.97,
|
| 1844 |
+
"grad_norm": 1.9763495922088623,
|
| 1845 |
+
"learning_rate": 1.7035775127768313e-06,
|
| 1846 |
+
"loss": 0.0539,
|
| 1847 |
+
"step": 2330
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 3.99,
|
| 1851 |
+
"grad_norm": 0.11824575811624527,
|
| 1852 |
+
"learning_rate": 8.517887563884157e-07,
|
| 1853 |
+
"loss": 0.0322,
|
| 1854 |
+
"step": 2340
|
| 1855 |
+
},
|
| 1856 |
{
|
| 1857 |
"epoch": 4.0,
|
| 1858 |
+
"step": 2348,
|
| 1859 |
+
"total_flos": 2.910419581971751e+18,
|
| 1860 |
+
"train_loss": 0.4888155373286145,
|
| 1861 |
+
"train_runtime": 2894.9609,
|
| 1862 |
+
"train_samples_per_second": 12.973,
|
| 1863 |
+
"train_steps_per_second": 0.811
|
| 1864 |
}
|
| 1865 |
],
|
| 1866 |
"logging_steps": 10,
|
| 1867 |
+
"max_steps": 2348,
|
| 1868 |
"num_input_tokens_seen": 0,
|
| 1869 |
"num_train_epochs": 4,
|
| 1870 |
"save_steps": 100,
|
| 1871 |
+
"total_flos": 2.910419581971751e+18,
|
| 1872 |
"train_batch_size": 16,
|
| 1873 |
"trial_name": null,
|
| 1874 |
"trial_params": null
|