End of training
Browse files- README.md +18 -4
- all_results.json +10 -10
- eval_results.json +5 -5
- log_history.json +1319 -1137
- train_results.json +5 -5
- trainer_state.json +1324 -1142
README.md
CHANGED
|
@@ -1,6 +1,7 @@
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
tags:
|
|
|
|
| 4 |
- generated_from_trainer
|
| 5 |
datasets:
|
| 6 |
- voxceleb
|
|
@@ -8,7 +9,20 @@ metrics:
|
|
| 8 |
- accuracy
|
| 9 |
model-index:
|
| 10 |
- name: ecapa-tdnn-voxceleb1-c512-aam
|
| 11 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
---
|
| 13 |
|
| 14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -16,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# ecapa-tdnn-voxceleb1-c512-aam
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [](https://huggingface.co/) on the voxceleb dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss:
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
tags:
|
| 4 |
+
- audio-classification
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- voxceleb
|
|
|
|
| 9 |
- accuracy
|
| 10 |
model-index:
|
| 11 |
- name: ecapa-tdnn-voxceleb1-c512-aam
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Audio Classification
|
| 15 |
+
type: audio-classification
|
| 16 |
+
dataset:
|
| 17 |
+
name: confit/voxceleb
|
| 18 |
+
type: voxceleb
|
| 19 |
+
config: verification
|
| 20 |
+
split: train
|
| 21 |
+
args: verification
|
| 22 |
+
metrics:
|
| 23 |
+
- name: Accuracy
|
| 24 |
+
type: accuracy
|
| 25 |
+
value: 0.9757901815736382
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 30 |
|
| 31 |
# ecapa-tdnn-voxceleb1-c512-aam
|
| 32 |
|
| 33 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on the confit/voxceleb dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.5840
|
| 36 |
+
- Accuracy: 0.9758
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos": 2.
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9757901815736382,
|
| 4 |
+
"eval_loss": 0.5840117335319519,
|
| 5 |
+
"eval_runtime": 13.1656,
|
| 6 |
+
"eval_samples_per_second": 112.946,
|
| 7 |
+
"eval_steps_per_second": 112.946,
|
| 8 |
+
"total_flos": 2.7398100529152e+18,
|
| 9 |
+
"train_loss": 2.9414075751926587,
|
| 10 |
+
"train_runtime": 59857.6179,
|
| 11 |
+
"train_samples_per_second": 24.584,
|
| 12 |
+
"train_steps_per_second": 0.096
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9757901815736382,
|
| 4 |
+
"eval_loss": 0.5840117335319519,
|
| 5 |
+
"eval_runtime": 13.1656,
|
| 6 |
+
"eval_samples_per_second": 112.946,
|
| 7 |
+
"eval_steps_per_second": 112.946
|
| 8 |
}
|
log_history.json
CHANGED
|
@@ -1,1928 +1,2110 @@
|
|
| 1 |
[
|
| 2 |
{
|
| 3 |
-
"loss": 13.
|
| 4 |
-
"grad_norm": 6.
|
| 5 |
-
"learning_rate":
|
| 6 |
-
"epoch": 0.
|
| 7 |
"step": 20
|
| 8 |
},
|
| 9 |
{
|
| 10 |
-
"loss": 13.
|
| 11 |
-
"grad_norm":
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"epoch": 0.
|
| 14 |
"step": 40
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"loss": 13.
|
| 18 |
-
"grad_norm":
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"epoch": 0.
|
| 21 |
"step": 60
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"loss":
|
| 25 |
-
"grad_norm":
|
| 26 |
-
"learning_rate":
|
| 27 |
-
"epoch": 0.
|
| 28 |
"step": 80
|
| 29 |
},
|
| 30 |
{
|
| 31 |
-
"loss":
|
| 32 |
-
"grad_norm":
|
| 33 |
-
"learning_rate":
|
| 34 |
-
"epoch": 0.
|
| 35 |
"step": 100
|
| 36 |
},
|
| 37 |
{
|
| 38 |
-
"loss": 12.
|
| 39 |
-
"grad_norm":
|
| 40 |
-
"learning_rate":
|
| 41 |
-
"epoch": 0.
|
| 42 |
"step": 120
|
| 43 |
},
|
| 44 |
{
|
| 45 |
-
"loss": 12.
|
| 46 |
-
"grad_norm":
|
| 47 |
-
"learning_rate":
|
| 48 |
-
"epoch": 0.
|
| 49 |
"step": 140
|
| 50 |
},
|
| 51 |
{
|
| 52 |
-
"loss": 12.
|
| 53 |
-
"grad_norm":
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"epoch": 0.
|
| 56 |
"step": 160
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"loss": 12.
|
| 60 |
-
"grad_norm":
|
| 61 |
-
"learning_rate":
|
| 62 |
-
"epoch": 0.
|
| 63 |
"step": 180
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"loss":
|
| 67 |
-
"grad_norm":
|
| 68 |
-
"learning_rate":
|
| 69 |
-
"epoch": 0.
|
| 70 |
"step": 200
|
| 71 |
},
|
| 72 |
{
|
| 73 |
-
"loss":
|
| 74 |
-
"grad_norm":
|
| 75 |
-
"learning_rate":
|
| 76 |
-
"epoch": 0.
|
| 77 |
"step": 220
|
| 78 |
},
|
| 79 |
{
|
| 80 |
-
"loss":
|
| 81 |
-
"grad_norm":
|
| 82 |
-
"learning_rate":
|
| 83 |
-
"epoch": 0.
|
| 84 |
"step": 240
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"loss":
|
| 88 |
-
"grad_norm":
|
| 89 |
-
"learning_rate":
|
| 90 |
-
"epoch": 0.
|
| 91 |
"step": 260
|
| 92 |
},
|
| 93 |
{
|
| 94 |
-
"loss":
|
| 95 |
-
"grad_norm":
|
| 96 |
-
"learning_rate":
|
| 97 |
-
"epoch": 0.
|
| 98 |
"step": 280
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"loss":
|
| 102 |
-
"grad_norm":
|
| 103 |
-
"learning_rate":
|
| 104 |
-
"epoch": 0.
|
| 105 |
"step": 300
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"loss":
|
| 109 |
-
"grad_norm": 3.
|
| 110 |
-
"learning_rate":
|
| 111 |
-
"epoch": 0.
|
| 112 |
"step": 320
|
| 113 |
},
|
| 114 |
{
|
| 115 |
-
"loss":
|
| 116 |
-
"grad_norm": 3.
|
| 117 |
-
"learning_rate":
|
| 118 |
-
"epoch": 0.
|
| 119 |
"step": 340
|
| 120 |
},
|
| 121 |
{
|
| 122 |
-
"loss":
|
| 123 |
-
"grad_norm": 3.
|
| 124 |
-
"learning_rate":
|
| 125 |
-
"epoch": 0.
|
| 126 |
"step": 360
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"loss":
|
| 130 |
-
"grad_norm": 3.
|
| 131 |
-
"learning_rate":
|
| 132 |
-
"epoch": 0.
|
| 133 |
"step": 380
|
| 134 |
},
|
| 135 |
{
|
| 136 |
-
"loss":
|
| 137 |
-
"grad_norm": 3.
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"epoch": 0.
|
| 140 |
"step": 400
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"loss":
|
| 144 |
-
"grad_norm": 3.
|
| 145 |
-
"learning_rate":
|
| 146 |
-
"epoch": 0.
|
| 147 |
"step": 420
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"loss":
|
| 151 |
-
"grad_norm": 3.
|
| 152 |
-
"learning_rate":
|
| 153 |
-
"epoch": 0.
|
| 154 |
"step": 440
|
| 155 |
},
|
| 156 |
{
|
| 157 |
-
"loss":
|
| 158 |
-
"grad_norm": 3.
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"epoch": 0.
|
| 161 |
"step": 460
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"loss":
|
| 165 |
-
"grad_norm": 3.
|
| 166 |
-
"learning_rate":
|
| 167 |
-
"epoch": 0.
|
| 168 |
"step": 480
|
| 169 |
},
|
| 170 |
{
|
| 171 |
-
"loss":
|
| 172 |
-
"grad_norm": 3.
|
| 173 |
-
"learning_rate":
|
| 174 |
-
"epoch": 0.
|
| 175 |
"step": 500
|
| 176 |
},
|
| 177 |
{
|
| 178 |
-
"loss":
|
| 179 |
-
"grad_norm": 3.
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"epoch": 0.
|
| 182 |
"step": 520
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"
|
| 186 |
-
"
|
| 187 |
-
"
|
| 188 |
-
"
|
| 189 |
-
"eval_steps_per_second": 25.083,
|
| 190 |
-
"epoch": 1.0,
|
| 191 |
-
"step": 523
|
| 192 |
-
},
|
| 193 |
-
{
|
| 194 |
-
"loss": 11.2589,
|
| 195 |
-
"grad_norm": 3.6838159561157227,
|
| 196 |
-
"learning_rate": 9.963883577650308e-05,
|
| 197 |
-
"epoch": 1.0325047801147227,
|
| 198 |
"step": 540
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"loss":
|
| 202 |
-
"grad_norm": 3.
|
| 203 |
-
"learning_rate":
|
| 204 |
-
"epoch":
|
| 205 |
"step": 560
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"
|
| 209 |
-
"
|
| 210 |
-
"
|
| 211 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 212 |
"step": 580
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"loss":
|
| 216 |
-
"grad_norm": 3.
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"epoch": 1.
|
| 219 |
"step": 600
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"loss":
|
| 223 |
-
"grad_norm": 3.
|
| 224 |
-
"learning_rate":
|
| 225 |
-
"epoch": 1.
|
| 226 |
"step": 620
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"loss":
|
| 230 |
-
"grad_norm": 3.
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"epoch": 1.
|
| 233 |
"step": 640
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"loss":
|
| 237 |
-
"grad_norm": 3.
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"epoch": 1.
|
| 240 |
"step": 660
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"loss":
|
| 244 |
-
"grad_norm": 3.
|
| 245 |
-
"learning_rate":
|
| 246 |
-
"epoch": 1.
|
| 247 |
"step": 680
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"loss":
|
| 251 |
-
"grad_norm": 3.
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"epoch": 1.
|
| 254 |
"step": 700
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"loss":
|
| 258 |
-
"grad_norm": 3.
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"epoch": 1.
|
| 261 |
"step": 720
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"loss":
|
| 265 |
-
"grad_norm":
|
| 266 |
-
"learning_rate":
|
| 267 |
-
"epoch": 1.
|
| 268 |
"step": 740
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"loss":
|
| 272 |
-
"grad_norm": 3.
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"epoch": 1.
|
| 275 |
"step": 760
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"loss":
|
| 279 |
-
"grad_norm": 3.
|
| 280 |
-
"learning_rate":
|
| 281 |
-
"epoch": 1.
|
| 282 |
"step": 780
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"loss":
|
| 286 |
-
"grad_norm": 3.
|
| 287 |
-
"learning_rate":
|
| 288 |
-
"epoch": 1.
|
| 289 |
"step": 800
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"loss":
|
| 293 |
-
"grad_norm":
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"epoch": 1.
|
| 296 |
"step": 820
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"loss":
|
| 300 |
-
"grad_norm":
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"epoch": 1.
|
| 303 |
"step": 840
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"loss":
|
| 307 |
-
"grad_norm":
|
| 308 |
-
"learning_rate":
|
| 309 |
-
"epoch": 1.
|
| 310 |
"step": 860
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"loss":
|
| 314 |
-
"grad_norm":
|
| 315 |
-
"learning_rate":
|
| 316 |
-
"epoch": 1.
|
| 317 |
"step": 880
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"loss":
|
| 321 |
-
"grad_norm":
|
| 322 |
-
"learning_rate":
|
| 323 |
-
"epoch": 1.
|
| 324 |
"step": 900
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"loss":
|
| 328 |
-
"grad_norm":
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"epoch": 1.
|
| 331 |
"step": 920
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"loss":
|
| 335 |
-
"grad_norm":
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"epoch": 1.
|
| 338 |
"step": 940
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"loss":
|
| 342 |
-
"grad_norm":
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"epoch": 1.
|
| 345 |
"step": 960
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"loss":
|
| 349 |
-
"grad_norm":
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"epoch": 1.
|
| 352 |
"step": 980
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"loss":
|
| 356 |
-
"grad_norm":
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"epoch": 1.
|
| 359 |
"step": 1000
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"loss":
|
| 363 |
-
"grad_norm":
|
| 364 |
-
"learning_rate":
|
| 365 |
-
"epoch": 1.
|
| 366 |
"step": 1020
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"loss":
|
| 370 |
-
"grad_norm":
|
| 371 |
-
"learning_rate":
|
| 372 |
-
"epoch": 1.
|
| 373 |
"step": 1040
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"
|
| 377 |
-
"
|
| 378 |
-
"
|
| 379 |
-
"
|
| 380 |
-
"eval_steps_per_second": 32.226,
|
| 381 |
-
"epoch": 2.0,
|
| 382 |
-
"step": 1046
|
| 383 |
-
},
|
| 384 |
-
{
|
| 385 |
-
"loss": 9.714,
|
| 386 |
-
"grad_norm": 4.428006172180176,
|
| 387 |
-
"learning_rate": 8.859145952836202e-05,
|
| 388 |
-
"epoch": 2.026768642447419,
|
| 389 |
"step": 1060
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"loss":
|
| 393 |
-
"grad_norm":
|
| 394 |
-
"learning_rate":
|
| 395 |
-
"epoch":
|
| 396 |
"step": 1080
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"loss":
|
| 400 |
-
"grad_norm": 4.
|
| 401 |
-
"learning_rate":
|
| 402 |
-
"epoch":
|
| 403 |
"step": 1100
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"loss":
|
| 407 |
-
"grad_norm":
|
| 408 |
-
"learning_rate":
|
| 409 |
-
"epoch":
|
| 410 |
"step": 1120
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"loss":
|
| 414 |
-
"grad_norm": 4.
|
| 415 |
-
"learning_rate":
|
| 416 |
-
"epoch":
|
| 417 |
"step": 1140
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"
|
| 421 |
-
"
|
| 422 |
-
"
|
| 423 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 424 |
"step": 1160
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"loss":
|
| 428 |
-
"grad_norm":
|
| 429 |
-
"learning_rate":
|
| 430 |
-
"epoch": 2.
|
| 431 |
"step": 1180
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"loss":
|
| 435 |
-
"grad_norm": 4.
|
| 436 |
-
"learning_rate":
|
| 437 |
-
"epoch": 2.
|
| 438 |
"step": 1200
|
| 439 |
},
|
| 440 |
{
|
| 441 |
-
"loss":
|
| 442 |
-
"grad_norm": 4.
|
| 443 |
-
"learning_rate":
|
| 444 |
-
"epoch": 2.
|
| 445 |
"step": 1220
|
| 446 |
},
|
| 447 |
{
|
| 448 |
-
"loss":
|
| 449 |
-
"grad_norm": 4.
|
| 450 |
-
"learning_rate":
|
| 451 |
-
"epoch": 2.
|
| 452 |
"step": 1240
|
| 453 |
},
|
| 454 |
{
|
| 455 |
-
"loss":
|
| 456 |
-
"grad_norm": 4.
|
| 457 |
-
"learning_rate":
|
| 458 |
-
"epoch": 2.
|
| 459 |
"step": 1260
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"loss":
|
| 463 |
-
"grad_norm": 4.
|
| 464 |
-
"learning_rate":
|
| 465 |
-
"epoch": 2.
|
| 466 |
"step": 1280
|
| 467 |
},
|
| 468 |
{
|
| 469 |
-
"loss":
|
| 470 |
-
"grad_norm":
|
| 471 |
-
"learning_rate":
|
| 472 |
-
"epoch": 2.
|
| 473 |
"step": 1300
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"loss":
|
| 477 |
-
"grad_norm": 4.
|
| 478 |
-
"learning_rate":
|
| 479 |
-
"epoch": 2.
|
| 480 |
"step": 1320
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"loss":
|
| 484 |
-
"grad_norm": 4.
|
| 485 |
-
"learning_rate":
|
| 486 |
-
"epoch": 2.
|
| 487 |
"step": 1340
|
| 488 |
},
|
| 489 |
{
|
| 490 |
-
"loss":
|
| 491 |
-
"grad_norm": 4.
|
| 492 |
-
"learning_rate":
|
| 493 |
-
"epoch": 2.
|
| 494 |
"step": 1360
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"loss":
|
| 498 |
-
"grad_norm": 4.
|
| 499 |
-
"learning_rate":
|
| 500 |
-
"epoch": 2.
|
| 501 |
"step": 1380
|
| 502 |
},
|
| 503 |
{
|
| 504 |
-
"loss":
|
| 505 |
-
"grad_norm": 4.
|
| 506 |
-
"learning_rate":
|
| 507 |
-
"epoch": 2.
|
| 508 |
"step": 1400
|
| 509 |
},
|
| 510 |
{
|
| 511 |
-
"loss":
|
| 512 |
-
"grad_norm": 4.
|
| 513 |
-
"learning_rate":
|
| 514 |
-
"epoch": 2.
|
| 515 |
"step": 1420
|
| 516 |
},
|
| 517 |
{
|
| 518 |
-
"loss":
|
| 519 |
-
"grad_norm": 4.
|
| 520 |
-
"learning_rate":
|
| 521 |
-
"epoch": 2.
|
| 522 |
"step": 1440
|
| 523 |
},
|
| 524 |
{
|
| 525 |
-
"loss":
|
| 526 |
-
"grad_norm": 4.
|
| 527 |
-
"learning_rate":
|
| 528 |
-
"epoch": 2.
|
| 529 |
"step": 1460
|
| 530 |
},
|
| 531 |
{
|
| 532 |
-
"loss":
|
| 533 |
-
"grad_norm": 4.
|
| 534 |
-
"learning_rate":
|
| 535 |
-
"epoch": 2.
|
| 536 |
"step": 1480
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"loss":
|
| 540 |
-
"grad_norm": 4.
|
| 541 |
-
"learning_rate":
|
| 542 |
-
"epoch": 2.
|
| 543 |
"step": 1500
|
| 544 |
},
|
| 545 |
{
|
| 546 |
-
"loss":
|
| 547 |
-
"grad_norm": 4.
|
| 548 |
-
"learning_rate":
|
| 549 |
-
"epoch": 2.
|
| 550 |
"step": 1520
|
| 551 |
},
|
| 552 |
{
|
| 553 |
-
"loss":
|
| 554 |
-
"grad_norm":
|
| 555 |
-
"learning_rate":
|
| 556 |
-
"epoch": 2.
|
| 557 |
"step": 1540
|
| 558 |
},
|
| 559 |
{
|
| 560 |
-
"loss":
|
| 561 |
-
"grad_norm":
|
| 562 |
-
"learning_rate":
|
| 563 |
-
"epoch": 2.
|
| 564 |
"step": 1560
|
| 565 |
},
|
| 566 |
{
|
| 567 |
-
"
|
| 568 |
-
"
|
| 569 |
-
"
|
| 570 |
-
"
|
| 571 |
-
"eval_steps_per_second": 35.513,
|
| 572 |
-
"epoch": 3.0,
|
| 573 |
-
"step": 1569
|
| 574 |
-
},
|
| 575 |
-
{
|
| 576 |
-
"loss": 8.6104,
|
| 577 |
-
"grad_norm": 4.895749092102051,
|
| 578 |
-
"learning_rate": 7.754408328022095e-05,
|
| 579 |
-
"epoch": 3.0210325047801145,
|
| 580 |
"step": 1580
|
| 581 |
},
|
| 582 |
{
|
| 583 |
-
"loss":
|
| 584 |
-
"grad_norm":
|
| 585 |
-
"learning_rate":
|
| 586 |
-
"epoch":
|
| 587 |
"step": 1600
|
| 588 |
},
|
| 589 |
{
|
| 590 |
-
"loss":
|
| 591 |
-
"grad_norm":
|
| 592 |
-
"learning_rate":
|
| 593 |
-
"epoch":
|
| 594 |
"step": 1620
|
| 595 |
},
|
| 596 |
{
|
| 597 |
-
"loss":
|
| 598 |
-
"grad_norm":
|
| 599 |
-
"learning_rate":
|
| 600 |
-
"epoch":
|
| 601 |
"step": 1640
|
| 602 |
},
|
| 603 |
{
|
| 604 |
-
"loss":
|
| 605 |
-
"grad_norm":
|
| 606 |
-
"learning_rate":
|
| 607 |
-
"epoch":
|
| 608 |
"step": 1660
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"loss":
|
| 612 |
-
"grad_norm":
|
| 613 |
-
"learning_rate":
|
| 614 |
-
"epoch":
|
| 615 |
"step": 1680
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"loss":
|
| 619 |
-
"grad_norm":
|
| 620 |
-
"learning_rate":
|
| 621 |
-
"epoch":
|
| 622 |
"step": 1700
|
| 623 |
},
|
| 624 |
{
|
| 625 |
-
"loss":
|
| 626 |
-
"grad_norm":
|
| 627 |
-
"learning_rate":
|
| 628 |
-
"epoch":
|
| 629 |
"step": 1720
|
| 630 |
},
|
| 631 |
{
|
| 632 |
-
"
|
| 633 |
-
"
|
| 634 |
-
"
|
| 635 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 636 |
"step": 1740
|
| 637 |
},
|
| 638 |
{
|
| 639 |
-
"loss":
|
| 640 |
-
"grad_norm":
|
| 641 |
-
"learning_rate":
|
| 642 |
-
"epoch": 3.
|
| 643 |
"step": 1760
|
| 644 |
},
|
| 645 |
{
|
| 646 |
-
"loss":
|
| 647 |
-
"grad_norm":
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"epoch": 3.
|
| 650 |
"step": 1780
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"loss":
|
| 654 |
-
"grad_norm":
|
| 655 |
-
"learning_rate":
|
| 656 |
-
"epoch": 3.
|
| 657 |
"step": 1800
|
| 658 |
},
|
| 659 |
{
|
| 660 |
-
"loss":
|
| 661 |
-
"grad_norm":
|
| 662 |
-
"learning_rate":
|
| 663 |
-
"epoch": 3.
|
| 664 |
"step": 1820
|
| 665 |
},
|
| 666 |
{
|
| 667 |
-
"loss":
|
| 668 |
-
"grad_norm":
|
| 669 |
-
"learning_rate":
|
| 670 |
-
"epoch": 3.
|
| 671 |
"step": 1840
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"loss":
|
| 675 |
-
"grad_norm":
|
| 676 |
-
"learning_rate":
|
| 677 |
-
"epoch": 3.
|
| 678 |
"step": 1860
|
| 679 |
},
|
| 680 |
{
|
| 681 |
-
"loss":
|
| 682 |
-
"grad_norm":
|
| 683 |
-
"learning_rate":
|
| 684 |
-
"epoch": 3.
|
| 685 |
"step": 1880
|
| 686 |
},
|
| 687 |
{
|
| 688 |
-
"loss":
|
| 689 |
-
"grad_norm":
|
| 690 |
-
"learning_rate":
|
| 691 |
-
"epoch": 3.
|
| 692 |
"step": 1900
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"loss":
|
| 696 |
-
"grad_norm":
|
| 697 |
-
"learning_rate":
|
| 698 |
-
"epoch": 3.
|
| 699 |
"step": 1920
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"loss":
|
| 703 |
-
"grad_norm":
|
| 704 |
-
"learning_rate":
|
| 705 |
-
"epoch": 3.
|
| 706 |
"step": 1940
|
| 707 |
},
|
| 708 |
{
|
| 709 |
-
"loss":
|
| 710 |
-
"grad_norm":
|
| 711 |
-
"learning_rate":
|
| 712 |
-
"epoch": 3.
|
| 713 |
"step": 1960
|
| 714 |
},
|
| 715 |
{
|
| 716 |
-
"loss":
|
| 717 |
-
"grad_norm":
|
| 718 |
-
"learning_rate":
|
| 719 |
-
"epoch": 3.
|
| 720 |
"step": 1980
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"loss":
|
| 724 |
-
"grad_norm":
|
| 725 |
-
"learning_rate":
|
| 726 |
-
"epoch": 3.
|
| 727 |
"step": 2000
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"loss":
|
| 731 |
-
"grad_norm":
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"epoch": 3.
|
| 734 |
"step": 2020
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"loss":
|
| 738 |
-
"grad_norm":
|
| 739 |
-
"learning_rate":
|
| 740 |
-
"epoch": 3.
|
| 741 |
"step": 2040
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"loss":
|
| 745 |
-
"grad_norm":
|
| 746 |
-
"learning_rate":
|
| 747 |
-
"epoch": 3.
|
| 748 |
"step": 2060
|
| 749 |
},
|
| 750 |
{
|
| 751 |
-
"loss":
|
| 752 |
-
"grad_norm":
|
| 753 |
-
"learning_rate":
|
| 754 |
-
"epoch": 3.
|
| 755 |
"step": 2080
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"
|
| 759 |
-
"
|
| 760 |
-
"
|
| 761 |
-
"
|
| 762 |
-
"eval_steps_per_second": 40.482,
|
| 763 |
-
"epoch": 4.0,
|
| 764 |
-
"step": 2092
|
| 765 |
-
},
|
| 766 |
-
{
|
| 767 |
-
"loss": 7.8607,
|
| 768 |
-
"grad_norm": 5.820804595947266,
|
| 769 |
-
"learning_rate": 6.649670703207989e-05,
|
| 770 |
-
"epoch": 4.015296367112811,
|
| 771 |
"step": 2100
|
| 772 |
},
|
| 773 |
{
|
| 774 |
-
"loss":
|
| 775 |
-
"grad_norm":
|
| 776 |
-
"learning_rate":
|
| 777 |
-
"epoch":
|
| 778 |
"step": 2120
|
| 779 |
},
|
| 780 |
{
|
| 781 |
-
"loss":
|
| 782 |
-
"grad_norm":
|
| 783 |
-
"learning_rate":
|
| 784 |
-
"epoch":
|
| 785 |
"step": 2140
|
| 786 |
},
|
| 787 |
{
|
| 788 |
-
"loss":
|
| 789 |
-
"grad_norm":
|
| 790 |
-
"learning_rate":
|
| 791 |
-
"epoch":
|
| 792 |
"step": 2160
|
| 793 |
},
|
| 794 |
{
|
| 795 |
-
"loss":
|
| 796 |
-
"grad_norm":
|
| 797 |
-
"learning_rate":
|
| 798 |
-
"epoch":
|
| 799 |
"step": 2180
|
| 800 |
},
|
| 801 |
{
|
| 802 |
-
"loss":
|
| 803 |
-
"grad_norm":
|
| 804 |
-
"learning_rate":
|
| 805 |
-
"epoch":
|
| 806 |
"step": 2200
|
| 807 |
},
|
| 808 |
{
|
| 809 |
-
"loss":
|
| 810 |
-
"grad_norm":
|
| 811 |
-
"learning_rate":
|
| 812 |
-
"epoch":
|
| 813 |
"step": 2220
|
| 814 |
},
|
| 815 |
{
|
| 816 |
-
"loss":
|
| 817 |
-
"grad_norm":
|
| 818 |
-
"learning_rate":
|
| 819 |
-
"epoch":
|
| 820 |
"step": 2240
|
| 821 |
},
|
| 822 |
{
|
| 823 |
-
"loss":
|
| 824 |
-
"grad_norm":
|
| 825 |
-
"learning_rate":
|
| 826 |
-
"epoch":
|
| 827 |
"step": 2260
|
| 828 |
},
|
| 829 |
{
|
| 830 |
-
"loss":
|
| 831 |
-
"grad_norm":
|
| 832 |
-
"learning_rate":
|
| 833 |
-
"epoch":
|
| 834 |
"step": 2280
|
| 835 |
},
|
| 836 |
{
|
| 837 |
-
"loss":
|
| 838 |
-
"grad_norm":
|
| 839 |
-
"learning_rate":
|
| 840 |
-
"epoch": 4.
|
| 841 |
"step": 2300
|
| 842 |
},
|
| 843 |
{
|
| 844 |
-
"
|
| 845 |
-
"
|
| 846 |
-
"
|
| 847 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 848 |
"step": 2320
|
| 849 |
},
|
| 850 |
{
|
| 851 |
-
"loss":
|
| 852 |
-
"grad_norm":
|
| 853 |
-
"learning_rate":
|
| 854 |
-
"epoch": 4.
|
| 855 |
"step": 2340
|
| 856 |
},
|
| 857 |
{
|
| 858 |
-
"loss":
|
| 859 |
-
"grad_norm":
|
| 860 |
-
"learning_rate":
|
| 861 |
-
"epoch": 4.
|
| 862 |
"step": 2360
|
| 863 |
},
|
| 864 |
{
|
| 865 |
-
"loss":
|
| 866 |
-
"grad_norm":
|
| 867 |
-
"learning_rate":
|
| 868 |
-
"epoch": 4.
|
| 869 |
"step": 2380
|
| 870 |
},
|
| 871 |
{
|
| 872 |
-
"loss":
|
| 873 |
-
"grad_norm":
|
| 874 |
-
"learning_rate":
|
| 875 |
-
"epoch": 4.
|
| 876 |
"step": 2400
|
| 877 |
},
|
| 878 |
{
|
| 879 |
-
"loss":
|
| 880 |
-
"grad_norm":
|
| 881 |
-
"learning_rate":
|
| 882 |
-
"epoch": 4.
|
| 883 |
"step": 2420
|
| 884 |
},
|
| 885 |
{
|
| 886 |
-
"loss":
|
| 887 |
-
"grad_norm":
|
| 888 |
-
"learning_rate":
|
| 889 |
-
"epoch": 4.
|
| 890 |
"step": 2440
|
| 891 |
},
|
| 892 |
{
|
| 893 |
-
"loss":
|
| 894 |
-
"grad_norm":
|
| 895 |
-
"learning_rate":
|
| 896 |
-
"epoch": 4.
|
| 897 |
"step": 2460
|
| 898 |
},
|
| 899 |
{
|
| 900 |
-
"loss":
|
| 901 |
-
"grad_norm":
|
| 902 |
-
"learning_rate":
|
| 903 |
-
"epoch": 4.
|
| 904 |
"step": 2480
|
| 905 |
},
|
| 906 |
{
|
| 907 |
-
"loss":
|
| 908 |
-
"grad_norm":
|
| 909 |
-
"learning_rate":
|
| 910 |
-
"epoch": 4.
|
| 911 |
"step": 2500
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"loss":
|
| 915 |
-
"grad_norm":
|
| 916 |
-
"learning_rate":
|
| 917 |
-
"epoch": 4.
|
| 918 |
"step": 2520
|
| 919 |
},
|
| 920 |
{
|
| 921 |
-
"loss":
|
| 922 |
-
"grad_norm":
|
| 923 |
-
"learning_rate":
|
| 924 |
-
"epoch": 4.
|
| 925 |
"step": 2540
|
| 926 |
},
|
| 927 |
{
|
| 928 |
-
"loss":
|
| 929 |
-
"grad_norm":
|
| 930 |
-
"learning_rate":
|
| 931 |
-
"epoch": 4.
|
| 932 |
"step": 2560
|
| 933 |
},
|
| 934 |
{
|
| 935 |
-
"loss":
|
| 936 |
-
"grad_norm":
|
| 937 |
-
"learning_rate":
|
| 938 |
-
"epoch": 4.
|
| 939 |
"step": 2580
|
| 940 |
},
|
| 941 |
{
|
| 942 |
-
"loss":
|
| 943 |
-
"grad_norm":
|
| 944 |
-
"learning_rate":
|
| 945 |
-
"epoch": 4.
|
| 946 |
"step": 2600
|
| 947 |
},
|
| 948 |
{
|
| 949 |
-
"
|
| 950 |
-
"
|
| 951 |
-
"
|
| 952 |
-
"
|
| 953 |
-
"eval_steps_per_second": 194.433,
|
| 954 |
-
"epoch": 5.0,
|
| 955 |
-
"step": 2615
|
| 956 |
-
},
|
| 957 |
-
{
|
| 958 |
-
"loss": 7.1863,
|
| 959 |
-
"grad_norm": 6.838705062866211,
|
| 960 |
-
"learning_rate": 5.544933078393881e-05,
|
| 961 |
-
"epoch": 5.009560229445507,
|
| 962 |
"step": 2620
|
| 963 |
},
|
| 964 |
{
|
| 965 |
-
"loss":
|
| 966 |
-
"grad_norm":
|
| 967 |
-
"learning_rate":
|
| 968 |
-
"epoch":
|
| 969 |
"step": 2640
|
| 970 |
},
|
| 971 |
{
|
| 972 |
-
"loss":
|
| 973 |
-
"grad_norm":
|
| 974 |
-
"learning_rate":
|
| 975 |
-
"epoch":
|
| 976 |
"step": 2660
|
| 977 |
},
|
| 978 |
{
|
| 979 |
-
"loss":
|
| 980 |
-
"grad_norm":
|
| 981 |
-
"learning_rate":
|
| 982 |
-
"epoch":
|
| 983 |
"step": 2680
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"loss":
|
| 987 |
-
"grad_norm":
|
| 988 |
-
"learning_rate":
|
| 989 |
-
"epoch":
|
| 990 |
"step": 2700
|
| 991 |
},
|
| 992 |
{
|
| 993 |
-
"loss":
|
| 994 |
-
"grad_norm":
|
| 995 |
-
"learning_rate":
|
| 996 |
-
"epoch":
|
| 997 |
"step": 2720
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
-
"loss":
|
| 1001 |
-
"grad_norm":
|
| 1002 |
-
"learning_rate":
|
| 1003 |
-
"epoch":
|
| 1004 |
"step": 2740
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
-
"loss":
|
| 1008 |
-
"grad_norm":
|
| 1009 |
-
"learning_rate":
|
| 1010 |
-
"epoch":
|
| 1011 |
"step": 2760
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
-
"loss":
|
| 1015 |
-
"grad_norm":
|
| 1016 |
-
"learning_rate":
|
| 1017 |
-
"epoch":
|
| 1018 |
"step": 2780
|
| 1019 |
},
|
| 1020 |
{
|
| 1021 |
-
"loss":
|
| 1022 |
-
"grad_norm":
|
| 1023 |
-
"learning_rate":
|
| 1024 |
-
"epoch":
|
| 1025 |
"step": 2800
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
-
"loss":
|
| 1029 |
-
"grad_norm":
|
| 1030 |
-
"learning_rate":
|
| 1031 |
-
"epoch":
|
| 1032 |
"step": 2820
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
-
"loss":
|
| 1036 |
-
"grad_norm":
|
| 1037 |
-
"learning_rate":
|
| 1038 |
-
"epoch":
|
| 1039 |
"step": 2840
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
-
"loss":
|
| 1043 |
-
"grad_norm":
|
| 1044 |
-
"learning_rate":
|
| 1045 |
-
"epoch":
|
| 1046 |
"step": 2860
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
-
"
|
| 1050 |
-
"
|
| 1051 |
-
"
|
| 1052 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1053 |
"step": 2880
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
-
"loss":
|
| 1057 |
-
"grad_norm":
|
| 1058 |
-
"learning_rate":
|
| 1059 |
-
"epoch": 5.
|
| 1060 |
"step": 2900
|
| 1061 |
},
|
| 1062 |
{
|
| 1063 |
-
"loss":
|
| 1064 |
-
"grad_norm":
|
| 1065 |
-
"learning_rate":
|
| 1066 |
-
"epoch": 5.
|
| 1067 |
"step": 2920
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
-
"loss":
|
| 1071 |
-
"grad_norm":
|
| 1072 |
-
"learning_rate":
|
| 1073 |
-
"epoch": 5.
|
| 1074 |
"step": 2940
|
| 1075 |
},
|
| 1076 |
{
|
| 1077 |
-
"loss":
|
| 1078 |
-
"grad_norm":
|
| 1079 |
-
"learning_rate":
|
| 1080 |
-
"epoch": 5.
|
| 1081 |
"step": 2960
|
| 1082 |
},
|
| 1083 |
{
|
| 1084 |
-
"loss":
|
| 1085 |
-
"grad_norm":
|
| 1086 |
-
"learning_rate":
|
| 1087 |
-
"epoch": 5.
|
| 1088 |
"step": 2980
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
-
"loss":
|
| 1092 |
-
"grad_norm":
|
| 1093 |
-
"learning_rate":
|
| 1094 |
-
"epoch": 5.
|
| 1095 |
"step": 3000
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
-
"loss":
|
| 1099 |
-
"grad_norm":
|
| 1100 |
-
"learning_rate":
|
| 1101 |
-
"epoch": 5.
|
| 1102 |
"step": 3020
|
| 1103 |
},
|
| 1104 |
{
|
| 1105 |
-
"loss":
|
| 1106 |
-
"grad_norm":
|
| 1107 |
-
"learning_rate":
|
| 1108 |
-
"epoch": 5.
|
| 1109 |
"step": 3040
|
| 1110 |
},
|
| 1111 |
{
|
| 1112 |
-
"loss":
|
| 1113 |
-
"grad_norm":
|
| 1114 |
-
"learning_rate":
|
| 1115 |
-
"epoch": 5.
|
| 1116 |
"step": 3060
|
| 1117 |
},
|
| 1118 |
{
|
| 1119 |
-
"loss":
|
| 1120 |
-
"grad_norm":
|
| 1121 |
-
"learning_rate":
|
| 1122 |
-
"epoch": 5.
|
| 1123 |
"step": 3080
|
| 1124 |
},
|
| 1125 |
{
|
| 1126 |
-
"loss":
|
| 1127 |
-
"grad_norm":
|
| 1128 |
-
"learning_rate":
|
| 1129 |
-
"epoch": 5.
|
| 1130 |
"step": 3100
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
-
"loss":
|
| 1134 |
-
"grad_norm":
|
| 1135 |
-
"learning_rate":
|
| 1136 |
-
"epoch": 5.
|
| 1137 |
"step": 3120
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
-
"
|
| 1141 |
-
"
|
| 1142 |
-
"
|
| 1143 |
-
"
|
| 1144 |
-
"eval_steps_per_second": 33.48,
|
| 1145 |
-
"epoch": 6.0,
|
| 1146 |
-
"step": 3138
|
| 1147 |
-
},
|
| 1148 |
-
{
|
| 1149 |
-
"loss": 6.6927,
|
| 1150 |
-
"grad_norm": 7.1014723777771,
|
| 1151 |
-
"learning_rate": 4.440195453579775e-05,
|
| 1152 |
-
"epoch": 6.003824091778203,
|
| 1153 |
"step": 3140
|
| 1154 |
},
|
| 1155 |
{
|
| 1156 |
-
"loss":
|
| 1157 |
-
"grad_norm":
|
| 1158 |
-
"learning_rate":
|
| 1159 |
-
"epoch":
|
| 1160 |
"step": 3160
|
| 1161 |
},
|
| 1162 |
{
|
| 1163 |
-
"loss":
|
| 1164 |
-
"grad_norm":
|
| 1165 |
-
"learning_rate":
|
| 1166 |
-
"epoch":
|
| 1167 |
"step": 3180
|
| 1168 |
},
|
| 1169 |
{
|
| 1170 |
-
"loss":
|
| 1171 |
-
"grad_norm":
|
| 1172 |
-
"learning_rate":
|
| 1173 |
-
"epoch":
|
| 1174 |
"step": 3200
|
| 1175 |
},
|
| 1176 |
{
|
| 1177 |
-
"loss":
|
| 1178 |
-
"grad_norm":
|
| 1179 |
-
"learning_rate":
|
| 1180 |
-
"epoch": 6
|
| 1181 |
"step": 3220
|
| 1182 |
},
|
| 1183 |
{
|
| 1184 |
-
"loss":
|
| 1185 |
-
"grad_norm":
|
| 1186 |
-
"learning_rate":
|
| 1187 |
-
"epoch":
|
| 1188 |
"step": 3240
|
| 1189 |
},
|
| 1190 |
{
|
| 1191 |
-
"loss":
|
| 1192 |
-
"grad_norm":
|
| 1193 |
-
"learning_rate":
|
| 1194 |
-
"epoch":
|
| 1195 |
"step": 3260
|
| 1196 |
},
|
| 1197 |
{
|
| 1198 |
-
"loss":
|
| 1199 |
-
"grad_norm":
|
| 1200 |
-
"learning_rate":
|
| 1201 |
-
"epoch":
|
| 1202 |
"step": 3280
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
-
"loss":
|
| 1206 |
-
"grad_norm":
|
| 1207 |
-
"learning_rate":
|
| 1208 |
-
"epoch":
|
| 1209 |
"step": 3300
|
| 1210 |
},
|
| 1211 |
{
|
| 1212 |
-
"loss":
|
| 1213 |
-
"grad_norm":
|
| 1214 |
-
"learning_rate":
|
| 1215 |
-
"epoch":
|
| 1216 |
"step": 3320
|
| 1217 |
},
|
| 1218 |
{
|
| 1219 |
-
"loss":
|
| 1220 |
-
"grad_norm":
|
| 1221 |
-
"learning_rate":
|
| 1222 |
-
"epoch":
|
| 1223 |
"step": 3340
|
| 1224 |
},
|
| 1225 |
{
|
| 1226 |
-
"loss":
|
| 1227 |
-
"grad_norm":
|
| 1228 |
-
"learning_rate":
|
| 1229 |
-
"epoch":
|
| 1230 |
"step": 3360
|
| 1231 |
},
|
| 1232 |
{
|
| 1233 |
-
"loss":
|
| 1234 |
-
"grad_norm":
|
| 1235 |
-
"learning_rate":
|
| 1236 |
-
"epoch":
|
| 1237 |
"step": 3380
|
| 1238 |
},
|
| 1239 |
{
|
| 1240 |
-
"loss":
|
| 1241 |
-
"grad_norm":
|
| 1242 |
-
"learning_rate":
|
| 1243 |
-
"epoch":
|
| 1244 |
"step": 3400
|
| 1245 |
},
|
| 1246 |
{
|
| 1247 |
-
"loss":
|
| 1248 |
-
"grad_norm":
|
| 1249 |
-
"learning_rate":
|
| 1250 |
-
"epoch":
|
| 1251 |
"step": 3420
|
| 1252 |
},
|
| 1253 |
{
|
| 1254 |
-
"loss":
|
| 1255 |
-
"grad_norm":
|
| 1256 |
-
"learning_rate":
|
| 1257 |
-
"epoch":
|
| 1258 |
"step": 3440
|
| 1259 |
},
|
| 1260 |
{
|
| 1261 |
-
"
|
| 1262 |
-
"
|
| 1263 |
-
"
|
| 1264 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1265 |
"step": 3460
|
| 1266 |
},
|
| 1267 |
{
|
| 1268 |
-
"loss":
|
| 1269 |
-
"grad_norm":
|
| 1270 |
-
"learning_rate":
|
| 1271 |
-
"epoch": 6.
|
| 1272 |
"step": 3480
|
| 1273 |
},
|
| 1274 |
{
|
| 1275 |
-
"loss":
|
| 1276 |
-
"grad_norm":
|
| 1277 |
-
"learning_rate":
|
| 1278 |
-
"epoch": 6.
|
| 1279 |
"step": 3500
|
| 1280 |
},
|
| 1281 |
{
|
| 1282 |
-
"loss":
|
| 1283 |
-
"grad_norm":
|
| 1284 |
-
"learning_rate":
|
| 1285 |
-
"epoch": 6.
|
| 1286 |
"step": 3520
|
| 1287 |
},
|
| 1288 |
{
|
| 1289 |
-
"loss":
|
| 1290 |
-
"grad_norm":
|
| 1291 |
-
"learning_rate":
|
| 1292 |
-
"epoch": 6.
|
| 1293 |
"step": 3540
|
| 1294 |
},
|
| 1295 |
{
|
| 1296 |
-
"loss":
|
| 1297 |
-
"grad_norm":
|
| 1298 |
-
"learning_rate":
|
| 1299 |
-
"epoch": 6.
|
| 1300 |
"step": 3560
|
| 1301 |
},
|
| 1302 |
{
|
| 1303 |
-
"loss":
|
| 1304 |
-
"grad_norm":
|
| 1305 |
-
"learning_rate":
|
| 1306 |
-
"epoch": 6.
|
| 1307 |
"step": 3580
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
-
"loss":
|
| 1311 |
-
"grad_norm":
|
| 1312 |
-
"learning_rate":
|
| 1313 |
-
"epoch": 6.
|
| 1314 |
"step": 3600
|
| 1315 |
},
|
| 1316 |
{
|
| 1317 |
-
"loss":
|
| 1318 |
-
"grad_norm":
|
| 1319 |
-
"learning_rate":
|
| 1320 |
-
"epoch": 6.
|
| 1321 |
"step": 3620
|
| 1322 |
},
|
| 1323 |
{
|
| 1324 |
-
"loss":
|
| 1325 |
-
"grad_norm":
|
| 1326 |
-
"learning_rate":
|
| 1327 |
-
"epoch": 6.
|
| 1328 |
"step": 3640
|
| 1329 |
},
|
| 1330 |
{
|
| 1331 |
-
"loss":
|
| 1332 |
-
"grad_norm":
|
| 1333 |
-
"learning_rate":
|
| 1334 |
-
"epoch": 6.
|
| 1335 |
"step": 3660
|
| 1336 |
},
|
| 1337 |
{
|
| 1338 |
-
"
|
| 1339 |
-
"
|
| 1340 |
-
"
|
| 1341 |
-
"
|
| 1342 |
-
"eval_steps_per_second": 30.688,
|
| 1343 |
-
"epoch": 7.0,
|
| 1344 |
-
"step": 3661
|
| 1345 |
-
},
|
| 1346 |
-
{
|
| 1347 |
-
"loss": 6.1764,
|
| 1348 |
-
"grad_norm": 7.424711227416992,
|
| 1349 |
-
"learning_rate": 3.292967920118972e-05,
|
| 1350 |
-
"epoch": 7.036328871892925,
|
| 1351 |
"step": 3680
|
| 1352 |
},
|
| 1353 |
{
|
| 1354 |
-
"loss":
|
| 1355 |
-
"grad_norm":
|
| 1356 |
-
"learning_rate":
|
| 1357 |
-
"epoch":
|
| 1358 |
"step": 3700
|
| 1359 |
},
|
| 1360 |
{
|
| 1361 |
-
"loss":
|
| 1362 |
-
"grad_norm":
|
| 1363 |
-
"learning_rate":
|
| 1364 |
-
"epoch":
|
| 1365 |
"step": 3720
|
| 1366 |
},
|
| 1367 |
{
|
| 1368 |
-
"loss":
|
| 1369 |
-
"grad_norm":
|
| 1370 |
-
"learning_rate":
|
| 1371 |
-
"epoch":
|
| 1372 |
"step": 3740
|
| 1373 |
},
|
| 1374 |
{
|
| 1375 |
-
"loss":
|
| 1376 |
-
"grad_norm":
|
| 1377 |
-
"learning_rate":
|
| 1378 |
-
"epoch":
|
| 1379 |
"step": 3760
|
| 1380 |
},
|
| 1381 |
{
|
| 1382 |
-
"loss":
|
| 1383 |
-
"grad_norm":
|
| 1384 |
-
"learning_rate":
|
| 1385 |
-
"epoch":
|
| 1386 |
"step": 3780
|
| 1387 |
},
|
| 1388 |
{
|
| 1389 |
-
"loss":
|
| 1390 |
-
"grad_norm":
|
| 1391 |
-
"learning_rate":
|
| 1392 |
-
"epoch":
|
| 1393 |
"step": 3800
|
| 1394 |
},
|
| 1395 |
{
|
| 1396 |
-
"loss":
|
| 1397 |
-
"grad_norm":
|
| 1398 |
-
"learning_rate":
|
| 1399 |
-
"epoch":
|
| 1400 |
"step": 3820
|
| 1401 |
},
|
| 1402 |
{
|
| 1403 |
-
"loss":
|
| 1404 |
-
"grad_norm":
|
| 1405 |
-
"learning_rate":
|
| 1406 |
-
"epoch":
|
| 1407 |
"step": 3840
|
| 1408 |
},
|
| 1409 |
{
|
| 1410 |
-
"loss":
|
| 1411 |
-
"grad_norm":
|
| 1412 |
-
"learning_rate":
|
| 1413 |
-
"epoch":
|
| 1414 |
"step": 3860
|
| 1415 |
},
|
| 1416 |
{
|
| 1417 |
-
"loss":
|
| 1418 |
-
"grad_norm":
|
| 1419 |
-
"learning_rate":
|
| 1420 |
-
"epoch":
|
| 1421 |
"step": 3880
|
| 1422 |
},
|
| 1423 |
{
|
| 1424 |
-
"loss":
|
| 1425 |
-
"grad_norm":
|
| 1426 |
-
"learning_rate":
|
| 1427 |
-
"epoch":
|
| 1428 |
"step": 3900
|
| 1429 |
},
|
| 1430 |
{
|
| 1431 |
-
"loss":
|
| 1432 |
-
"grad_norm":
|
| 1433 |
-
"learning_rate":
|
| 1434 |
-
"epoch":
|
| 1435 |
"step": 3920
|
| 1436 |
},
|
| 1437 |
{
|
| 1438 |
-
"loss":
|
| 1439 |
-
"grad_norm":
|
| 1440 |
-
"learning_rate":
|
| 1441 |
-
"epoch":
|
| 1442 |
"step": 3940
|
| 1443 |
},
|
| 1444 |
{
|
| 1445 |
-
"loss":
|
| 1446 |
-
"grad_norm":
|
| 1447 |
-
"learning_rate":
|
| 1448 |
-
"epoch":
|
| 1449 |
"step": 3960
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
-
"loss":
|
| 1453 |
-
"grad_norm":
|
| 1454 |
-
"learning_rate":
|
| 1455 |
-
"epoch":
|
| 1456 |
"step": 3980
|
| 1457 |
},
|
| 1458 |
{
|
| 1459 |
-
"loss":
|
| 1460 |
-
"grad_norm":
|
| 1461 |
-
"learning_rate":
|
| 1462 |
-
"epoch":
|
| 1463 |
"step": 4000
|
| 1464 |
},
|
| 1465 |
{
|
| 1466 |
-
"loss":
|
| 1467 |
-
"grad_norm":
|
| 1468 |
-
"learning_rate":
|
| 1469 |
-
"epoch":
|
| 1470 |
"step": 4020
|
| 1471 |
},
|
| 1472 |
{
|
| 1473 |
-
"
|
| 1474 |
-
"
|
| 1475 |
-
"
|
| 1476 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1477 |
"step": 4040
|
| 1478 |
},
|
| 1479 |
{
|
| 1480 |
-
"loss":
|
| 1481 |
-
"grad_norm":
|
| 1482 |
-
"learning_rate":
|
| 1483 |
-
"epoch": 7.
|
| 1484 |
"step": 4060
|
| 1485 |
},
|
| 1486 |
{
|
| 1487 |
-
"loss":
|
| 1488 |
-
"grad_norm":
|
| 1489 |
-
"learning_rate":
|
| 1490 |
-
"epoch": 7.
|
| 1491 |
"step": 4080
|
| 1492 |
},
|
| 1493 |
{
|
| 1494 |
-
"loss":
|
| 1495 |
-
"grad_norm":
|
| 1496 |
-
"learning_rate":
|
| 1497 |
-
"epoch": 7.
|
| 1498 |
"step": 4100
|
| 1499 |
},
|
| 1500 |
{
|
| 1501 |
-
"loss":
|
| 1502 |
-
"grad_norm":
|
| 1503 |
-
"learning_rate":
|
| 1504 |
-
"epoch": 7.
|
| 1505 |
"step": 4120
|
| 1506 |
},
|
| 1507 |
{
|
| 1508 |
-
"loss":
|
| 1509 |
-
"grad_norm":
|
| 1510 |
-
"learning_rate":
|
| 1511 |
-
"epoch": 7.
|
| 1512 |
"step": 4140
|
| 1513 |
},
|
| 1514 |
{
|
| 1515 |
-
"loss":
|
| 1516 |
-
"grad_norm":
|
| 1517 |
-
"learning_rate":
|
| 1518 |
-
"epoch": 7.
|
| 1519 |
"step": 4160
|
| 1520 |
},
|
| 1521 |
{
|
| 1522 |
-
"loss":
|
| 1523 |
-
"grad_norm":
|
| 1524 |
-
"learning_rate":
|
| 1525 |
-
"epoch": 7.
|
| 1526 |
"step": 4180
|
| 1527 |
},
|
| 1528 |
{
|
| 1529 |
-
"
|
| 1530 |
-
"
|
| 1531 |
-
"
|
| 1532 |
-
"
|
| 1533 |
-
"eval_steps_per_second": 57.14,
|
| 1534 |
-
"epoch": 8.0,
|
| 1535 |
-
"step": 4184
|
| 1536 |
-
},
|
| 1537 |
-
{
|
| 1538 |
-
"loss": 6.0763,
|
| 1539 |
-
"grad_norm": 7.686340808868408,
|
| 1540 |
-
"learning_rate": 2.1882302953048652e-05,
|
| 1541 |
-
"epoch": 8.030592734225621,
|
| 1542 |
"step": 4200
|
| 1543 |
},
|
| 1544 |
{
|
| 1545 |
-
"loss":
|
| 1546 |
-
"grad_norm":
|
| 1547 |
-
"learning_rate":
|
| 1548 |
-
"epoch":
|
| 1549 |
"step": 4220
|
| 1550 |
},
|
| 1551 |
{
|
| 1552 |
-
"loss":
|
| 1553 |
-
"grad_norm":
|
| 1554 |
-
"learning_rate":
|
| 1555 |
-
"epoch":
|
| 1556 |
"step": 4240
|
| 1557 |
},
|
| 1558 |
{
|
| 1559 |
-
"loss":
|
| 1560 |
-
"grad_norm":
|
| 1561 |
-
"learning_rate":
|
| 1562 |
-
"epoch":
|
| 1563 |
"step": 4260
|
| 1564 |
},
|
| 1565 |
{
|
| 1566 |
-
"loss":
|
| 1567 |
-
"grad_norm":
|
| 1568 |
-
"learning_rate":
|
| 1569 |
-
"epoch":
|
| 1570 |
"step": 4280
|
| 1571 |
},
|
| 1572 |
{
|
| 1573 |
-
"loss":
|
| 1574 |
-
"grad_norm":
|
| 1575 |
-
"learning_rate":
|
| 1576 |
-
"epoch":
|
| 1577 |
"step": 4300
|
| 1578 |
},
|
| 1579 |
{
|
| 1580 |
-
"loss":
|
| 1581 |
-
"grad_norm":
|
| 1582 |
-
"learning_rate":
|
| 1583 |
-
"epoch":
|
| 1584 |
"step": 4320
|
| 1585 |
},
|
| 1586 |
{
|
| 1587 |
-
"loss":
|
| 1588 |
-
"grad_norm":
|
| 1589 |
-
"learning_rate":
|
| 1590 |
-
"epoch":
|
| 1591 |
"step": 4340
|
| 1592 |
},
|
| 1593 |
{
|
| 1594 |
-
"loss":
|
| 1595 |
-
"grad_norm":
|
| 1596 |
-
"learning_rate":
|
| 1597 |
-
"epoch":
|
| 1598 |
"step": 4360
|
| 1599 |
},
|
| 1600 |
{
|
| 1601 |
-
"loss":
|
| 1602 |
-
"grad_norm":
|
| 1603 |
-
"learning_rate":
|
| 1604 |
-
"epoch":
|
| 1605 |
"step": 4380
|
| 1606 |
},
|
| 1607 |
{
|
| 1608 |
-
"loss":
|
| 1609 |
-
"grad_norm":
|
| 1610 |
-
"learning_rate":
|
| 1611 |
-
"epoch":
|
| 1612 |
"step": 4400
|
| 1613 |
},
|
| 1614 |
{
|
| 1615 |
-
"loss":
|
| 1616 |
-
"grad_norm":
|
| 1617 |
-
"learning_rate":
|
| 1618 |
-
"epoch":
|
| 1619 |
"step": 4420
|
| 1620 |
},
|
| 1621 |
{
|
| 1622 |
-
"loss":
|
| 1623 |
-
"grad_norm":
|
| 1624 |
-
"learning_rate":
|
| 1625 |
-
"epoch":
|
| 1626 |
"step": 4440
|
| 1627 |
},
|
| 1628 |
{
|
| 1629 |
-
"loss":
|
| 1630 |
-
"grad_norm":
|
| 1631 |
-
"learning_rate":
|
| 1632 |
-
"epoch":
|
| 1633 |
"step": 4460
|
| 1634 |
},
|
| 1635 |
{
|
| 1636 |
-
"loss":
|
| 1637 |
-
"grad_norm":
|
| 1638 |
-
"learning_rate":
|
| 1639 |
-
"epoch":
|
| 1640 |
"step": 4480
|
| 1641 |
},
|
| 1642 |
{
|
| 1643 |
-
"loss":
|
| 1644 |
-
"grad_norm":
|
| 1645 |
-
"learning_rate":
|
| 1646 |
-
"epoch":
|
| 1647 |
"step": 4500
|
| 1648 |
},
|
| 1649 |
{
|
| 1650 |
-
"loss":
|
| 1651 |
-
"grad_norm":
|
| 1652 |
-
"learning_rate":
|
| 1653 |
-
"epoch":
|
| 1654 |
"step": 4520
|
| 1655 |
},
|
| 1656 |
{
|
| 1657 |
-
"loss":
|
| 1658 |
-
"grad_norm":
|
| 1659 |
-
"learning_rate":
|
| 1660 |
-
"epoch":
|
| 1661 |
"step": 4540
|
| 1662 |
},
|
| 1663 |
{
|
| 1664 |
-
"loss":
|
| 1665 |
-
"grad_norm":
|
| 1666 |
-
"learning_rate":
|
| 1667 |
-
"epoch":
|
| 1668 |
"step": 4560
|
| 1669 |
},
|
| 1670 |
{
|
| 1671 |
-
"loss":
|
| 1672 |
-
"grad_norm":
|
| 1673 |
-
"learning_rate":
|
| 1674 |
-
"epoch":
|
| 1675 |
"step": 4580
|
| 1676 |
},
|
| 1677 |
{
|
| 1678 |
-
"loss":
|
| 1679 |
-
"grad_norm":
|
| 1680 |
-
"learning_rate":
|
| 1681 |
-
"epoch": 8.
|
| 1682 |
"step": 4600
|
| 1683 |
},
|
| 1684 |
{
|
| 1685 |
-
"
|
| 1686 |
-
"
|
| 1687 |
-
"
|
| 1688 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1689 |
"step": 4620
|
| 1690 |
},
|
| 1691 |
{
|
| 1692 |
-
"loss":
|
| 1693 |
-
"grad_norm":
|
| 1694 |
-
"learning_rate":
|
| 1695 |
-
"epoch": 8.
|
| 1696 |
"step": 4640
|
| 1697 |
},
|
| 1698 |
{
|
| 1699 |
-
"loss":
|
| 1700 |
-
"grad_norm":
|
| 1701 |
-
"learning_rate":
|
| 1702 |
-
"epoch": 8.
|
| 1703 |
"step": 4660
|
| 1704 |
},
|
| 1705 |
{
|
| 1706 |
-
"loss":
|
| 1707 |
-
"grad_norm":
|
| 1708 |
-
"learning_rate":
|
| 1709 |
-
"epoch": 8.
|
| 1710 |
"step": 4680
|
| 1711 |
},
|
| 1712 |
{
|
| 1713 |
-
"loss":
|
| 1714 |
-
"grad_norm":
|
| 1715 |
-
"learning_rate":
|
| 1716 |
-
"epoch": 8.
|
| 1717 |
"step": 4700
|
| 1718 |
},
|
| 1719 |
{
|
| 1720 |
-
"
|
| 1721 |
-
"
|
| 1722 |
-
"
|
| 1723 |
-
"
|
| 1724 |
-
"eval_steps_per_second": 27.985,
|
| 1725 |
-
"epoch": 9.0,
|
| 1726 |
-
"step": 4707
|
| 1727 |
-
},
|
| 1728 |
-
{
|
| 1729 |
-
"loss": 5.7978,
|
| 1730 |
-
"grad_norm": 7.77038049697876,
|
| 1731 |
-
"learning_rate": 1.0834926704907584e-05,
|
| 1732 |
-
"epoch": 9.024856596558317,
|
| 1733 |
"step": 4720
|
| 1734 |
},
|
| 1735 |
{
|
| 1736 |
-
"loss":
|
| 1737 |
-
"grad_norm":
|
| 1738 |
-
"learning_rate":
|
| 1739 |
-
"epoch":
|
| 1740 |
"step": 4740
|
| 1741 |
},
|
| 1742 |
{
|
| 1743 |
-
"loss":
|
| 1744 |
-
"grad_norm":
|
| 1745 |
-
"learning_rate":
|
| 1746 |
-
"epoch":
|
| 1747 |
"step": 4760
|
| 1748 |
},
|
| 1749 |
{
|
| 1750 |
-
"loss":
|
| 1751 |
-
"grad_norm":
|
| 1752 |
-
"learning_rate":
|
| 1753 |
-
"epoch":
|
| 1754 |
"step": 4780
|
| 1755 |
},
|
| 1756 |
{
|
| 1757 |
-
"loss":
|
| 1758 |
-
"grad_norm":
|
| 1759 |
-
"learning_rate":
|
| 1760 |
-
"epoch":
|
| 1761 |
"step": 4800
|
| 1762 |
},
|
| 1763 |
{
|
| 1764 |
-
"loss":
|
| 1765 |
-
"grad_norm":
|
| 1766 |
-
"learning_rate":
|
| 1767 |
-
"epoch":
|
| 1768 |
"step": 4820
|
| 1769 |
},
|
| 1770 |
{
|
| 1771 |
-
"loss":
|
| 1772 |
-
"grad_norm":
|
| 1773 |
-
"learning_rate":
|
| 1774 |
-
"epoch":
|
| 1775 |
"step": 4840
|
| 1776 |
},
|
| 1777 |
{
|
| 1778 |
-
"loss":
|
| 1779 |
-
"grad_norm":
|
| 1780 |
-
"learning_rate":
|
| 1781 |
-
"epoch":
|
| 1782 |
"step": 4860
|
| 1783 |
},
|
| 1784 |
{
|
| 1785 |
-
"loss":
|
| 1786 |
-
"grad_norm":
|
| 1787 |
-
"learning_rate":
|
| 1788 |
-
"epoch":
|
| 1789 |
"step": 4880
|
| 1790 |
},
|
| 1791 |
{
|
| 1792 |
-
"loss":
|
| 1793 |
-
"grad_norm":
|
| 1794 |
-
"learning_rate":
|
| 1795 |
-
"epoch":
|
| 1796 |
"step": 4900
|
| 1797 |
},
|
| 1798 |
{
|
| 1799 |
-
"loss":
|
| 1800 |
-
"grad_norm":
|
| 1801 |
-
"learning_rate":
|
| 1802 |
-
"epoch":
|
| 1803 |
"step": 4920
|
| 1804 |
},
|
| 1805 |
{
|
| 1806 |
-
"loss":
|
| 1807 |
-
"grad_norm":
|
| 1808 |
-
"learning_rate":
|
| 1809 |
-
"epoch":
|
| 1810 |
"step": 4940
|
| 1811 |
},
|
| 1812 |
{
|
| 1813 |
-
"loss":
|
| 1814 |
-
"grad_norm":
|
| 1815 |
-
"learning_rate":
|
| 1816 |
-
"epoch":
|
| 1817 |
"step": 4960
|
| 1818 |
},
|
| 1819 |
{
|
| 1820 |
-
"loss":
|
| 1821 |
-
"grad_norm":
|
| 1822 |
-
"learning_rate":
|
| 1823 |
-
"epoch":
|
| 1824 |
"step": 4980
|
| 1825 |
},
|
| 1826 |
{
|
| 1827 |
-
"loss":
|
| 1828 |
-
"grad_norm":
|
| 1829 |
-
"learning_rate":
|
| 1830 |
-
"epoch":
|
| 1831 |
"step": 5000
|
| 1832 |
},
|
| 1833 |
{
|
| 1834 |
-
"loss":
|
| 1835 |
-
"grad_norm":
|
| 1836 |
-
"learning_rate":
|
| 1837 |
-
"epoch":
|
| 1838 |
"step": 5020
|
| 1839 |
},
|
| 1840 |
{
|
| 1841 |
-
"loss":
|
| 1842 |
-
"grad_norm":
|
| 1843 |
-
"learning_rate":
|
| 1844 |
-
"epoch":
|
| 1845 |
"step": 5040
|
| 1846 |
},
|
| 1847 |
{
|
| 1848 |
-
"loss":
|
| 1849 |
-
"grad_norm":
|
| 1850 |
-
"learning_rate":
|
| 1851 |
-
"epoch":
|
| 1852 |
"step": 5060
|
| 1853 |
},
|
| 1854 |
{
|
| 1855 |
-
"loss":
|
| 1856 |
-
"grad_norm":
|
| 1857 |
-
"learning_rate":
|
| 1858 |
-
"epoch":
|
| 1859 |
"step": 5080
|
| 1860 |
},
|
| 1861 |
{
|
| 1862 |
-
"loss":
|
| 1863 |
-
"grad_norm":
|
| 1864 |
-
"learning_rate":
|
| 1865 |
-
"epoch":
|
| 1866 |
"step": 5100
|
| 1867 |
},
|
| 1868 |
{
|
| 1869 |
-
"loss":
|
| 1870 |
-
"grad_norm":
|
| 1871 |
-
"learning_rate":
|
| 1872 |
-
"epoch":
|
| 1873 |
"step": 5120
|
| 1874 |
},
|
| 1875 |
{
|
| 1876 |
-
"loss":
|
| 1877 |
-
"grad_norm":
|
| 1878 |
-
"learning_rate":
|
| 1879 |
-
"epoch":
|
| 1880 |
"step": 5140
|
| 1881 |
},
|
| 1882 |
{
|
| 1883 |
-
"loss":
|
| 1884 |
-
"grad_norm":
|
| 1885 |
-
"learning_rate":
|
| 1886 |
-
"epoch":
|
| 1887 |
"step": 5160
|
| 1888 |
},
|
| 1889 |
{
|
| 1890 |
-
"
|
| 1891 |
-
"
|
| 1892 |
-
"
|
| 1893 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1894 |
"step": 5180
|
| 1895 |
},
|
| 1896 |
{
|
| 1897 |
-
"loss":
|
| 1898 |
-
"grad_norm":
|
| 1899 |
-
"learning_rate":
|
| 1900 |
-
"epoch": 9.
|
| 1901 |
"step": 5200
|
| 1902 |
},
|
| 1903 |
{
|
| 1904 |
-
"loss":
|
| 1905 |
-
"grad_norm":
|
| 1906 |
-
"learning_rate":
|
| 1907 |
-
"epoch": 9.
|
| 1908 |
"step": 5220
|
| 1909 |
},
|
| 1910 |
{
|
| 1911 |
-
"
|
| 1912 |
-
"
|
| 1913 |
-
"
|
| 1914 |
-
"
|
| 1915 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1916 |
"epoch": 10.0,
|
| 1917 |
-
"step":
|
| 1918 |
},
|
| 1919 |
{
|
| 1920 |
-
"train_runtime":
|
| 1921 |
-
"train_samples_per_second":
|
| 1922 |
-
"train_steps_per_second": 0.
|
| 1923 |
-
"total_flos": 2.
|
| 1924 |
-
"train_loss":
|
| 1925 |
"epoch": 10.0,
|
| 1926 |
-
"step":
|
| 1927 |
}
|
| 1928 |
]
|
|
|
|
| 1 |
[
|
| 2 |
{
|
| 3 |
+
"loss": 13.2026,
|
| 4 |
+
"grad_norm": 6.155358791351318,
|
| 5 |
+
"learning_rate": 1.739130434782609e-05,
|
| 6 |
+
"epoch": 0.034782608695652174,
|
| 7 |
"step": 20
|
| 8 |
},
|
| 9 |
{
|
| 10 |
+
"loss": 13.1252,
|
| 11 |
+
"grad_norm": 5.816741943359375,
|
| 12 |
+
"learning_rate": 3.478260869565218e-05,
|
| 13 |
+
"epoch": 0.06956521739130435,
|
| 14 |
"step": 40
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"loss": 13.0001,
|
| 18 |
+
"grad_norm": 5.273156642913818,
|
| 19 |
+
"learning_rate": 5.2173913043478256e-05,
|
| 20 |
+
"epoch": 0.10434782608695652,
|
| 21 |
"step": 60
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"loss": 12.8639,
|
| 25 |
+
"grad_norm": 4.86655330657959,
|
| 26 |
+
"learning_rate": 6.956521739130436e-05,
|
| 27 |
+
"epoch": 0.1391304347826087,
|
| 28 |
"step": 80
|
| 29 |
},
|
| 30 |
{
|
| 31 |
+
"loss": 12.7376,
|
| 32 |
+
"grad_norm": 4.438321113586426,
|
| 33 |
+
"learning_rate": 8.695652173913044e-05,
|
| 34 |
+
"epoch": 0.17391304347826086,
|
| 35 |
"step": 100
|
| 36 |
},
|
| 37 |
{
|
| 38 |
+
"loss": 12.5722,
|
| 39 |
+
"grad_norm": 4.164404392242432,
|
| 40 |
+
"learning_rate": 0.00010434782608695651,
|
| 41 |
+
"epoch": 0.20869565217391303,
|
| 42 |
"step": 120
|
| 43 |
},
|
| 44 |
{
|
| 45 |
+
"loss": 12.4229,
|
| 46 |
+
"grad_norm": 3.858990430831909,
|
| 47 |
+
"learning_rate": 0.00012173913043478261,
|
| 48 |
+
"epoch": 0.24347826086956523,
|
| 49 |
"step": 140
|
| 50 |
},
|
| 51 |
{
|
| 52 |
+
"loss": 12.2581,
|
| 53 |
+
"grad_norm": 3.6574394702911377,
|
| 54 |
+
"learning_rate": 0.0001391304347826087,
|
| 55 |
+
"epoch": 0.2782608695652174,
|
| 56 |
"step": 160
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"loss": 12.0753,
|
| 60 |
+
"grad_norm": 3.3787951469421387,
|
| 61 |
+
"learning_rate": 0.0001565217391304348,
|
| 62 |
+
"epoch": 0.3130434782608696,
|
| 63 |
"step": 180
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"loss": 11.9261,
|
| 67 |
+
"grad_norm": 3.323820114135742,
|
| 68 |
+
"learning_rate": 0.00017391304347826088,
|
| 69 |
+
"epoch": 0.34782608695652173,
|
| 70 |
"step": 200
|
| 71 |
},
|
| 72 |
{
|
| 73 |
+
"loss": 11.7417,
|
| 74 |
+
"grad_norm": 3.247619152069092,
|
| 75 |
+
"learning_rate": 0.00019130434782608697,
|
| 76 |
+
"epoch": 0.3826086956521739,
|
| 77 |
"step": 220
|
| 78 |
},
|
| 79 |
{
|
| 80 |
+
"loss": 11.5771,
|
| 81 |
+
"grad_norm": 3.2254152297973633,
|
| 82 |
+
"learning_rate": 0.00020869565217391303,
|
| 83 |
+
"epoch": 0.41739130434782606,
|
| 84 |
"step": 240
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"loss": 11.3969,
|
| 88 |
+
"grad_norm": 3.1803464889526367,
|
| 89 |
+
"learning_rate": 0.00022608695652173914,
|
| 90 |
+
"epoch": 0.45217391304347826,
|
| 91 |
"step": 260
|
| 92 |
},
|
| 93 |
{
|
| 94 |
+
"loss": 11.2684,
|
| 95 |
+
"grad_norm": 3.41034197807312,
|
| 96 |
+
"learning_rate": 0.00024347826086956522,
|
| 97 |
+
"epoch": 0.48695652173913045,
|
| 98 |
"step": 280
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"loss": 11.0744,
|
| 102 |
+
"grad_norm": 3.246403217315674,
|
| 103 |
+
"learning_rate": 0.0002608695652173913,
|
| 104 |
+
"epoch": 0.5217391304347826,
|
| 105 |
"step": 300
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"loss": 10.8929,
|
| 109 |
+
"grad_norm": 3.202021360397339,
|
| 110 |
+
"learning_rate": 0.0002782608695652174,
|
| 111 |
+
"epoch": 0.5565217391304348,
|
| 112 |
"step": 320
|
| 113 |
},
|
| 114 |
{
|
| 115 |
+
"loss": 10.7468,
|
| 116 |
+
"grad_norm": 3.1231367588043213,
|
| 117 |
+
"learning_rate": 0.0002956521739130435,
|
| 118 |
+
"epoch": 0.591304347826087,
|
| 119 |
"step": 340
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"loss": 10.606,
|
| 123 |
+
"grad_norm": 3.1820390224456787,
|
| 124 |
+
"learning_rate": 0.0003130434782608696,
|
| 125 |
+
"epoch": 0.6260869565217392,
|
| 126 |
"step": 360
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"loss": 10.4871,
|
| 130 |
+
"grad_norm": 3.2470555305480957,
|
| 131 |
+
"learning_rate": 0.0003304347826086956,
|
| 132 |
+
"epoch": 0.6608695652173913,
|
| 133 |
"step": 380
|
| 134 |
},
|
| 135 |
{
|
| 136 |
+
"loss": 10.2836,
|
| 137 |
+
"grad_norm": 3.2452709674835205,
|
| 138 |
+
"learning_rate": 0.00034782608695652176,
|
| 139 |
+
"epoch": 0.6956521739130435,
|
| 140 |
"step": 400
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"loss": 10.1154,
|
| 144 |
+
"grad_norm": 3.203894853591919,
|
| 145 |
+
"learning_rate": 0.00036521739130434785,
|
| 146 |
+
"epoch": 0.7304347826086957,
|
| 147 |
"step": 420
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"loss": 9.9283,
|
| 151 |
+
"grad_norm": 3.269970178604126,
|
| 152 |
+
"learning_rate": 0.00038260869565217393,
|
| 153 |
+
"epoch": 0.7652173913043478,
|
| 154 |
"step": 440
|
| 155 |
},
|
| 156 |
{
|
| 157 |
+
"loss": 9.8674,
|
| 158 |
+
"grad_norm": 3.261357545852661,
|
| 159 |
+
"learning_rate": 0.0004,
|
| 160 |
+
"epoch": 0.8,
|
| 161 |
"step": 460
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"loss": 9.6224,
|
| 165 |
+
"grad_norm": 3.393953323364258,
|
| 166 |
+
"learning_rate": 0.00041739130434782605,
|
| 167 |
+
"epoch": 0.8347826086956521,
|
| 168 |
"step": 480
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"loss": 9.524,
|
| 172 |
+
"grad_norm": 3.321411609649658,
|
| 173 |
+
"learning_rate": 0.0004347826086956522,
|
| 174 |
+
"epoch": 0.8695652173913043,
|
| 175 |
"step": 500
|
| 176 |
},
|
| 177 |
{
|
| 178 |
+
"loss": 9.384,
|
| 179 |
+
"grad_norm": 3.3886823654174805,
|
| 180 |
+
"learning_rate": 0.0004521739130434783,
|
| 181 |
+
"epoch": 0.9043478260869565,
|
| 182 |
"step": 520
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"loss": 9.1767,
|
| 186 |
+
"grad_norm": 3.4735491275787354,
|
| 187 |
+
"learning_rate": 0.00046956521739130436,
|
| 188 |
+
"epoch": 0.9391304347826087,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 189 |
"step": 540
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"loss": 9.047,
|
| 193 |
+
"grad_norm": 3.416966676712036,
|
| 194 |
+
"learning_rate": 0.00048695652173913045,
|
| 195 |
+
"epoch": 0.9739130434782609,
|
| 196 |
"step": 560
|
| 197 |
},
|
| 198 |
{
|
| 199 |
+
"eval_loss": 8.366157531738281,
|
| 200 |
+
"eval_accuracy": 0.43039677202420984,
|
| 201 |
+
"eval_runtime": 42.3364,
|
| 202 |
+
"eval_samples_per_second": 35.123,
|
| 203 |
+
"eval_steps_per_second": 35.123,
|
| 204 |
+
"epoch": 1.0,
|
| 205 |
+
"step": 575
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"loss": 8.8835,
|
| 209 |
+
"grad_norm": 3.446899890899658,
|
| 210 |
+
"learning_rate": 0.0004995169082125604,
|
| 211 |
+
"epoch": 1.008695652173913,
|
| 212 |
"step": 580
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"loss": 8.6436,
|
| 216 |
+
"grad_norm": 3.5842247009277344,
|
| 217 |
+
"learning_rate": 0.0004975845410628019,
|
| 218 |
+
"epoch": 1.0434782608695652,
|
| 219 |
"step": 600
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"loss": 8.4775,
|
| 223 |
+
"grad_norm": 3.5029306411743164,
|
| 224 |
+
"learning_rate": 0.0004956521739130435,
|
| 225 |
+
"epoch": 1.0782608695652174,
|
| 226 |
"step": 620
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"loss": 8.322,
|
| 230 |
+
"grad_norm": 3.5451033115386963,
|
| 231 |
+
"learning_rate": 0.0004937198067632851,
|
| 232 |
+
"epoch": 1.1130434782608696,
|
| 233 |
"step": 640
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"loss": 8.1264,
|
| 237 |
+
"grad_norm": 3.5502634048461914,
|
| 238 |
+
"learning_rate": 0.0004917874396135266,
|
| 239 |
+
"epoch": 1.1478260869565218,
|
| 240 |
"step": 660
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"loss": 7.9905,
|
| 244 |
+
"grad_norm": 3.607395648956299,
|
| 245 |
+
"learning_rate": 0.0004898550724637681,
|
| 246 |
+
"epoch": 1.182608695652174,
|
| 247 |
"step": 680
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"loss": 7.8252,
|
| 251 |
+
"grad_norm": 3.6438565254211426,
|
| 252 |
+
"learning_rate": 0.0004879227053140097,
|
| 253 |
+
"epoch": 1.2173913043478262,
|
| 254 |
"step": 700
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"loss": 7.7737,
|
| 258 |
+
"grad_norm": 3.656705141067505,
|
| 259 |
+
"learning_rate": 0.0004859903381642512,
|
| 260 |
+
"epoch": 1.2521739130434781,
|
| 261 |
"step": 720
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"loss": 7.5822,
|
| 265 |
+
"grad_norm": 3.7424328327178955,
|
| 266 |
+
"learning_rate": 0.0004840579710144928,
|
| 267 |
+
"epoch": 1.2869565217391306,
|
| 268 |
"step": 740
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"loss": 7.4563,
|
| 272 |
+
"grad_norm": 3.673156261444092,
|
| 273 |
+
"learning_rate": 0.0004821256038647343,
|
| 274 |
+
"epoch": 1.3217391304347825,
|
| 275 |
"step": 760
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"loss": 7.3379,
|
| 279 |
+
"grad_norm": 3.6774067878723145,
|
| 280 |
+
"learning_rate": 0.0004801932367149758,
|
| 281 |
+
"epoch": 1.3565217391304347,
|
| 282 |
"step": 780
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"loss": 7.1559,
|
| 286 |
+
"grad_norm": 3.811283826828003,
|
| 287 |
+
"learning_rate": 0.0004782608695652174,
|
| 288 |
+
"epoch": 1.391304347826087,
|
| 289 |
"step": 800
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"loss": 7.0834,
|
| 293 |
+
"grad_norm": 3.7899839878082275,
|
| 294 |
+
"learning_rate": 0.00047632850241545894,
|
| 295 |
+
"epoch": 1.4260869565217391,
|
| 296 |
"step": 820
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"loss": 6.9172,
|
| 300 |
+
"grad_norm": 3.583247423171997,
|
| 301 |
+
"learning_rate": 0.00047439613526570047,
|
| 302 |
+
"epoch": 1.4608695652173913,
|
| 303 |
"step": 840
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"loss": 6.7251,
|
| 307 |
+
"grad_norm": 3.8192331790924072,
|
| 308 |
+
"learning_rate": 0.00047246376811594206,
|
| 309 |
+
"epoch": 1.4956521739130435,
|
| 310 |
"step": 860
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"loss": 6.7871,
|
| 314 |
+
"grad_norm": 3.8098299503326416,
|
| 315 |
+
"learning_rate": 0.0004705314009661836,
|
| 316 |
+
"epoch": 1.5304347826086957,
|
| 317 |
"step": 880
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"loss": 6.6103,
|
| 321 |
+
"grad_norm": 3.7341325283050537,
|
| 322 |
+
"learning_rate": 0.0004685990338164252,
|
| 323 |
+
"epoch": 1.5652173913043477,
|
| 324 |
"step": 900
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"loss": 6.4507,
|
| 328 |
+
"grad_norm": 3.9190495014190674,
|
| 329 |
+
"learning_rate": 0.00046666666666666666,
|
| 330 |
+
"epoch": 1.6,
|
| 331 |
"step": 920
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"loss": 6.3619,
|
| 335 |
+
"grad_norm": 3.9456422328948975,
|
| 336 |
+
"learning_rate": 0.0004647342995169082,
|
| 337 |
+
"epoch": 1.634782608695652,
|
| 338 |
"step": 940
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"loss": 6.2957,
|
| 342 |
+
"grad_norm": 3.899134874343872,
|
| 343 |
+
"learning_rate": 0.0004628019323671498,
|
| 344 |
+
"epoch": 1.6695652173913045,
|
| 345 |
"step": 960
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"loss": 6.1362,
|
| 349 |
+
"grad_norm": 3.878810167312622,
|
| 350 |
+
"learning_rate": 0.0004608695652173913,
|
| 351 |
+
"epoch": 1.7043478260869565,
|
| 352 |
"step": 980
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"loss": 5.9814,
|
| 356 |
+
"grad_norm": 3.9270784854888916,
|
| 357 |
+
"learning_rate": 0.00045893719806763285,
|
| 358 |
+
"epoch": 1.7391304347826086,
|
| 359 |
"step": 1000
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"loss": 5.9095,
|
| 363 |
+
"grad_norm": 3.8247644901275635,
|
| 364 |
+
"learning_rate": 0.00045700483091787444,
|
| 365 |
+
"epoch": 1.7739130434782608,
|
| 366 |
"step": 1020
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"loss": 5.7793,
|
| 370 |
+
"grad_norm": 3.8870134353637695,
|
| 371 |
+
"learning_rate": 0.000455072463768116,
|
| 372 |
+
"epoch": 1.808695652173913,
|
| 373 |
"step": 1040
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"loss": 5.7754,
|
| 377 |
+
"grad_norm": 3.9533441066741943,
|
| 378 |
+
"learning_rate": 0.00045314009661835745,
|
| 379 |
+
"epoch": 1.8434782608695652,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 380 |
"step": 1060
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"loss": 5.5886,
|
| 384 |
+
"grad_norm": 3.9928998947143555,
|
| 385 |
+
"learning_rate": 0.00045120772946859904,
|
| 386 |
+
"epoch": 1.8782608695652174,
|
| 387 |
"step": 1080
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"loss": 5.5482,
|
| 391 |
+
"grad_norm": 4.030064582824707,
|
| 392 |
+
"learning_rate": 0.0004492753623188406,
|
| 393 |
+
"epoch": 1.9130434782608696,
|
| 394 |
"step": 1100
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"loss": 5.4807,
|
| 398 |
+
"grad_norm": 3.961806297302246,
|
| 399 |
+
"learning_rate": 0.0004473429951690821,
|
| 400 |
+
"epoch": 1.9478260869565216,
|
| 401 |
"step": 1120
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"loss": 5.3508,
|
| 405 |
+
"grad_norm": 4.003119945526123,
|
| 406 |
+
"learning_rate": 0.0004454106280193237,
|
| 407 |
+
"epoch": 1.982608695652174,
|
| 408 |
"step": 1140
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"eval_loss": 4.025164604187012,
|
| 412 |
+
"eval_accuracy": 0.8190988567585743,
|
| 413 |
+
"eval_runtime": 42.7144,
|
| 414 |
+
"eval_samples_per_second": 34.813,
|
| 415 |
+
"eval_steps_per_second": 34.813,
|
| 416 |
+
"epoch": 2.0,
|
| 417 |
+
"step": 1150
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"loss": 5.1229,
|
| 421 |
+
"grad_norm": 3.958116292953491,
|
| 422 |
+
"learning_rate": 0.00044347826086956523,
|
| 423 |
+
"epoch": 2.017391304347826,
|
| 424 |
"step": 1160
|
| 425 |
},
|
| 426 |
{
|
| 427 |
+
"loss": 4.8146,
|
| 428 |
+
"grad_norm": 3.864279270172119,
|
| 429 |
+
"learning_rate": 0.00044154589371980677,
|
| 430 |
+
"epoch": 2.0521739130434784,
|
| 431 |
"step": 1180
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"loss": 4.8843,
|
| 435 |
+
"grad_norm": 4.045077323913574,
|
| 436 |
+
"learning_rate": 0.0004396135265700483,
|
| 437 |
+
"epoch": 2.0869565217391304,
|
| 438 |
"step": 1200
|
| 439 |
},
|
| 440 |
{
|
| 441 |
+
"loss": 4.8078,
|
| 442 |
+
"grad_norm": 4.061978816986084,
|
| 443 |
+
"learning_rate": 0.00043768115942028983,
|
| 444 |
+
"epoch": 2.121739130434783,
|
| 445 |
"step": 1220
|
| 446 |
},
|
| 447 |
{
|
| 448 |
+
"loss": 4.6812,
|
| 449 |
+
"grad_norm": 4.040159225463867,
|
| 450 |
+
"learning_rate": 0.0004357487922705314,
|
| 451 |
+
"epoch": 2.1565217391304348,
|
| 452 |
"step": 1240
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"loss": 4.6701,
|
| 456 |
+
"grad_norm": 4.234623908996582,
|
| 457 |
+
"learning_rate": 0.00043381642512077296,
|
| 458 |
+
"epoch": 2.1913043478260867,
|
| 459 |
"step": 1260
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"loss": 4.6221,
|
| 463 |
+
"grad_norm": 4.030038356781006,
|
| 464 |
+
"learning_rate": 0.0004318840579710145,
|
| 465 |
+
"epoch": 2.226086956521739,
|
| 466 |
"step": 1280
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"loss": 4.5647,
|
| 470 |
+
"grad_norm": 3.9954497814178467,
|
| 471 |
+
"learning_rate": 0.0004299516908212561,
|
| 472 |
+
"epoch": 2.260869565217391,
|
| 473 |
"step": 1300
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"loss": 4.4502,
|
| 477 |
+
"grad_norm": 4.188636779785156,
|
| 478 |
+
"learning_rate": 0.0004280193236714976,
|
| 479 |
+
"epoch": 2.2956521739130435,
|
| 480 |
"step": 1320
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"loss": 4.359,
|
| 484 |
+
"grad_norm": 4.185456275939941,
|
| 485 |
+
"learning_rate": 0.00042608695652173915,
|
| 486 |
+
"epoch": 2.3304347826086955,
|
| 487 |
"step": 1340
|
| 488 |
},
|
| 489 |
{
|
| 490 |
+
"loss": 4.2863,
|
| 491 |
+
"grad_norm": 4.123263359069824,
|
| 492 |
+
"learning_rate": 0.0004241545893719807,
|
| 493 |
+
"epoch": 2.365217391304348,
|
| 494 |
"step": 1360
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"loss": 4.3354,
|
| 498 |
+
"grad_norm": 4.194387435913086,
|
| 499 |
+
"learning_rate": 0.0004222222222222222,
|
| 500 |
+
"epoch": 2.4,
|
| 501 |
"step": 1380
|
| 502 |
},
|
| 503 |
{
|
| 504 |
+
"loss": 4.2176,
|
| 505 |
+
"grad_norm": 4.065763473510742,
|
| 506 |
+
"learning_rate": 0.00042028985507246375,
|
| 507 |
+
"epoch": 2.4347826086956523,
|
| 508 |
"step": 1400
|
| 509 |
},
|
| 510 |
{
|
| 511 |
+
"loss": 4.0597,
|
| 512 |
+
"grad_norm": 4.120363712310791,
|
| 513 |
+
"learning_rate": 0.00041835748792270534,
|
| 514 |
+
"epoch": 2.4695652173913043,
|
| 515 |
"step": 1420
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"loss": 4.028,
|
| 519 |
+
"grad_norm": 4.3197174072265625,
|
| 520 |
+
"learning_rate": 0.00041642512077294687,
|
| 521 |
+
"epoch": 2.5043478260869563,
|
| 522 |
"step": 1440
|
| 523 |
},
|
| 524 |
{
|
| 525 |
+
"loss": 3.9833,
|
| 526 |
+
"grad_norm": 4.2683610916137695,
|
| 527 |
+
"learning_rate": 0.0004144927536231884,
|
| 528 |
+
"epoch": 2.5391304347826087,
|
| 529 |
"step": 1460
|
| 530 |
},
|
| 531 |
{
|
| 532 |
+
"loss": 4.0065,
|
| 533 |
+
"grad_norm": 4.15448522567749,
|
| 534 |
+
"learning_rate": 0.00041256038647343,
|
| 535 |
+
"epoch": 2.573913043478261,
|
| 536 |
"step": 1480
|
| 537 |
},
|
| 538 |
{
|
| 539 |
+
"loss": 3.8134,
|
| 540 |
+
"grad_norm": 4.348177433013916,
|
| 541 |
+
"learning_rate": 0.0004106280193236715,
|
| 542 |
+
"epoch": 2.608695652173913,
|
| 543 |
"step": 1500
|
| 544 |
},
|
| 545 |
{
|
| 546 |
+
"loss": 3.8548,
|
| 547 |
+
"grad_norm": 4.100021839141846,
|
| 548 |
+
"learning_rate": 0.00040869565217391306,
|
| 549 |
+
"epoch": 2.643478260869565,
|
| 550 |
"step": 1520
|
| 551 |
},
|
| 552 |
{
|
| 553 |
+
"loss": 3.7814,
|
| 554 |
+
"grad_norm": 4.344174385070801,
|
| 555 |
+
"learning_rate": 0.0004067632850241546,
|
| 556 |
+
"epoch": 2.6782608695652175,
|
| 557 |
"step": 1540
|
| 558 |
},
|
| 559 |
{
|
| 560 |
+
"loss": 3.7578,
|
| 561 |
+
"grad_norm": 4.240079402923584,
|
| 562 |
+
"learning_rate": 0.00040483091787439613,
|
| 563 |
+
"epoch": 2.7130434782608694,
|
| 564 |
"step": 1560
|
| 565 |
},
|
| 566 |
{
|
| 567 |
+
"loss": 3.7331,
|
| 568 |
+
"grad_norm": 4.468689918518066,
|
| 569 |
+
"learning_rate": 0.0004028985507246377,
|
| 570 |
+
"epoch": 2.747826086956522,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 571 |
"step": 1580
|
| 572 |
},
|
| 573 |
{
|
| 574 |
+
"loss": 3.6396,
|
| 575 |
+
"grad_norm": 4.28464937210083,
|
| 576 |
+
"learning_rate": 0.00040096618357487925,
|
| 577 |
+
"epoch": 2.782608695652174,
|
| 578 |
"step": 1600
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"loss": 3.5799,
|
| 582 |
+
"grad_norm": 4.166805744171143,
|
| 583 |
+
"learning_rate": 0.0003990338164251208,
|
| 584 |
+
"epoch": 2.8173913043478263,
|
| 585 |
"step": 1620
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"loss": 3.4734,
|
| 589 |
+
"grad_norm": 4.237683296203613,
|
| 590 |
+
"learning_rate": 0.0003971014492753624,
|
| 591 |
+
"epoch": 2.8521739130434782,
|
| 592 |
"step": 1640
|
| 593 |
},
|
| 594 |
{
|
| 595 |
+
"loss": 3.5183,
|
| 596 |
+
"grad_norm": 4.153097152709961,
|
| 597 |
+
"learning_rate": 0.00039516908212560385,
|
| 598 |
+
"epoch": 2.8869565217391306,
|
| 599 |
"step": 1660
|
| 600 |
},
|
| 601 |
{
|
| 602 |
+
"loss": 3.3963,
|
| 603 |
+
"grad_norm": 4.2313947677612305,
|
| 604 |
+
"learning_rate": 0.0003932367149758454,
|
| 605 |
+
"epoch": 2.9217391304347826,
|
| 606 |
"step": 1680
|
| 607 |
},
|
| 608 |
{
|
| 609 |
+
"loss": 3.3081,
|
| 610 |
+
"grad_norm": 3.992475748062134,
|
| 611 |
+
"learning_rate": 0.000391304347826087,
|
| 612 |
+
"epoch": 2.9565217391304346,
|
| 613 |
"step": 1700
|
| 614 |
},
|
| 615 |
{
|
| 616 |
+
"loss": 3.3124,
|
| 617 |
+
"grad_norm": 4.4731059074401855,
|
| 618 |
+
"learning_rate": 0.0003893719806763285,
|
| 619 |
+
"epoch": 2.991304347826087,
|
| 620 |
"step": 1720
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"eval_loss": 2.1082653999328613,
|
| 624 |
+
"eval_accuracy": 0.9260255548083389,
|
| 625 |
+
"eval_runtime": 22.1676,
|
| 626 |
+
"eval_samples_per_second": 67.08,
|
| 627 |
+
"eval_steps_per_second": 67.08,
|
| 628 |
+
"epoch": 3.0,
|
| 629 |
+
"step": 1725
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"loss": 3.1247,
|
| 633 |
+
"grad_norm": 4.272000312805176,
|
| 634 |
+
"learning_rate": 0.00038743961352657004,
|
| 635 |
+
"epoch": 3.026086956521739,
|
| 636 |
"step": 1740
|
| 637 |
},
|
| 638 |
{
|
| 639 |
+
"loss": 3.1064,
|
| 640 |
+
"grad_norm": 4.102330207824707,
|
| 641 |
+
"learning_rate": 0.00038550724637681163,
|
| 642 |
+
"epoch": 3.0608695652173914,
|
| 643 |
"step": 1760
|
| 644 |
},
|
| 645 |
{
|
| 646 |
+
"loss": 2.9371,
|
| 647 |
+
"grad_norm": 4.381846904754639,
|
| 648 |
+
"learning_rate": 0.00038357487922705317,
|
| 649 |
+
"epoch": 3.0956521739130434,
|
| 650 |
"step": 1780
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"loss": 2.9355,
|
| 654 |
+
"grad_norm": 4.1588921546936035,
|
| 655 |
+
"learning_rate": 0.00038164251207729465,
|
| 656 |
+
"epoch": 3.130434782608696,
|
| 657 |
"step": 1800
|
| 658 |
},
|
| 659 |
{
|
| 660 |
+
"loss": 2.8545,
|
| 661 |
+
"grad_norm": 4.279609203338623,
|
| 662 |
+
"learning_rate": 0.00037971014492753623,
|
| 663 |
+
"epoch": 3.1652173913043478,
|
| 664 |
"step": 1820
|
| 665 |
},
|
| 666 |
{
|
| 667 |
+
"loss": 2.8096,
|
| 668 |
+
"grad_norm": 4.240756988525391,
|
| 669 |
+
"learning_rate": 0.00037777777777777777,
|
| 670 |
+
"epoch": 3.2,
|
| 671 |
"step": 1840
|
| 672 |
},
|
| 673 |
{
|
| 674 |
+
"loss": 2.8138,
|
| 675 |
+
"grad_norm": 4.11091947555542,
|
| 676 |
+
"learning_rate": 0.00037584541062801936,
|
| 677 |
+
"epoch": 3.234782608695652,
|
| 678 |
"step": 1860
|
| 679 |
},
|
| 680 |
{
|
| 681 |
+
"loss": 2.7417,
|
| 682 |
+
"grad_norm": 4.078794479370117,
|
| 683 |
+
"learning_rate": 0.0003739130434782609,
|
| 684 |
+
"epoch": 3.269565217391304,
|
| 685 |
"step": 1880
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"loss": 2.7937,
|
| 689 |
+
"grad_norm": 4.368116855621338,
|
| 690 |
+
"learning_rate": 0.0003719806763285024,
|
| 691 |
+
"epoch": 3.3043478260869565,
|
| 692 |
"step": 1900
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"loss": 2.7361,
|
| 696 |
+
"grad_norm": 4.044319152832031,
|
| 697 |
+
"learning_rate": 0.000370048309178744,
|
| 698 |
+
"epoch": 3.3391304347826085,
|
| 699 |
"step": 1920
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"loss": 2.7054,
|
| 703 |
+
"grad_norm": 4.314040184020996,
|
| 704 |
+
"learning_rate": 0.0003681159420289855,
|
| 705 |
+
"epoch": 3.373913043478261,
|
| 706 |
"step": 1940
|
| 707 |
},
|
| 708 |
{
|
| 709 |
+
"loss": 2.6682,
|
| 710 |
+
"grad_norm": 4.185855388641357,
|
| 711 |
+
"learning_rate": 0.000366183574879227,
|
| 712 |
+
"epoch": 3.408695652173913,
|
| 713 |
"step": 1960
|
| 714 |
},
|
| 715 |
{
|
| 716 |
+
"loss": 2.6644,
|
| 717 |
+
"grad_norm": 4.433622360229492,
|
| 718 |
+
"learning_rate": 0.0003642512077294686,
|
| 719 |
+
"epoch": 3.4434782608695653,
|
| 720 |
"step": 1980
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"loss": 2.618,
|
| 724 |
+
"grad_norm": 4.048947811126709,
|
| 725 |
+
"learning_rate": 0.00036231884057971015,
|
| 726 |
+
"epoch": 3.4782608695652173,
|
| 727 |
"step": 2000
|
| 728 |
},
|
| 729 |
{
|
| 730 |
+
"loss": 2.5982,
|
| 731 |
+
"grad_norm": 4.145406246185303,
|
| 732 |
+
"learning_rate": 0.0003603864734299517,
|
| 733 |
+
"epoch": 3.5130434782608697,
|
| 734 |
"step": 2020
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"loss": 2.6138,
|
| 738 |
+
"grad_norm": 4.2812910079956055,
|
| 739 |
+
"learning_rate": 0.00035845410628019327,
|
| 740 |
+
"epoch": 3.5478260869565217,
|
| 741 |
"step": 2040
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"loss": 2.5039,
|
| 745 |
+
"grad_norm": 4.400162220001221,
|
| 746 |
+
"learning_rate": 0.0003565217391304348,
|
| 747 |
+
"epoch": 3.5826086956521737,
|
| 748 |
"step": 2060
|
| 749 |
},
|
| 750 |
{
|
| 751 |
+
"loss": 2.5249,
|
| 752 |
+
"grad_norm": 4.217800617218018,
|
| 753 |
+
"learning_rate": 0.0003545893719806763,
|
| 754 |
+
"epoch": 3.617391304347826,
|
| 755 |
"step": 2080
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"loss": 2.4547,
|
| 759 |
+
"grad_norm": 4.076215744018555,
|
| 760 |
+
"learning_rate": 0.0003526570048309179,
|
| 761 |
+
"epoch": 3.6521739130434785,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 762 |
"step": 2100
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"loss": 2.4315,
|
| 766 |
+
"grad_norm": 4.139514446258545,
|
| 767 |
+
"learning_rate": 0.0003507246376811594,
|
| 768 |
+
"epoch": 3.6869565217391305,
|
| 769 |
"step": 2120
|
| 770 |
},
|
| 771 |
{
|
| 772 |
+
"loss": 2.3836,
|
| 773 |
+
"grad_norm": 4.118022918701172,
|
| 774 |
+
"learning_rate": 0.00034879227053140094,
|
| 775 |
+
"epoch": 3.7217391304347824,
|
| 776 |
"step": 2140
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"loss": 2.3284,
|
| 780 |
+
"grad_norm": 4.137601852416992,
|
| 781 |
+
"learning_rate": 0.00034685990338164253,
|
| 782 |
+
"epoch": 3.756521739130435,
|
| 783 |
"step": 2160
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"loss": 2.3095,
|
| 787 |
+
"grad_norm": 4.023979663848877,
|
| 788 |
+
"learning_rate": 0.00034492753623188406,
|
| 789 |
+
"epoch": 3.791304347826087,
|
| 790 |
"step": 2180
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"loss": 2.305,
|
| 794 |
+
"grad_norm": 4.042725086212158,
|
| 795 |
+
"learning_rate": 0.00034299516908212565,
|
| 796 |
+
"epoch": 3.8260869565217392,
|
| 797 |
"step": 2200
|
| 798 |
},
|
| 799 |
{
|
| 800 |
+
"loss": 2.3237,
|
| 801 |
+
"grad_norm": 4.265875339508057,
|
| 802 |
+
"learning_rate": 0.0003410628019323672,
|
| 803 |
+
"epoch": 3.860869565217391,
|
| 804 |
"step": 2220
|
| 805 |
},
|
| 806 |
{
|
| 807 |
+
"loss": 2.335,
|
| 808 |
+
"grad_norm": 4.205041408538818,
|
| 809 |
+
"learning_rate": 0.00033913043478260867,
|
| 810 |
+
"epoch": 3.8956521739130436,
|
| 811 |
"step": 2240
|
| 812 |
},
|
| 813 |
{
|
| 814 |
+
"loss": 2.2341,
|
| 815 |
+
"grad_norm": 4.1344709396362305,
|
| 816 |
+
"learning_rate": 0.00033719806763285025,
|
| 817 |
+
"epoch": 3.9304347826086956,
|
| 818 |
"step": 2260
|
| 819 |
},
|
| 820 |
{
|
| 821 |
+
"loss": 2.251,
|
| 822 |
+
"grad_norm": 4.247790813446045,
|
| 823 |
+
"learning_rate": 0.0003352657004830918,
|
| 824 |
+
"epoch": 3.965217391304348,
|
| 825 |
"step": 2280
|
| 826 |
},
|
| 827 |
{
|
| 828 |
+
"loss": 2.3212,
|
| 829 |
+
"grad_norm": 4.859626770019531,
|
| 830 |
+
"learning_rate": 0.0003333333333333333,
|
| 831 |
+
"epoch": 4.0,
|
| 832 |
"step": 2300
|
| 833 |
},
|
| 834 |
{
|
| 835 |
+
"eval_loss": 1.2223739624023438,
|
| 836 |
+
"eval_accuracy": 0.9435104236718225,
|
| 837 |
+
"eval_runtime": 14.8513,
|
| 838 |
+
"eval_samples_per_second": 100.126,
|
| 839 |
+
"eval_steps_per_second": 100.126,
|
| 840 |
+
"epoch": 4.0,
|
| 841 |
+
"step": 2300
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"loss": 1.9133,
|
| 845 |
+
"grad_norm": 4.098020553588867,
|
| 846 |
+
"learning_rate": 0.0003314009661835749,
|
| 847 |
+
"epoch": 4.034782608695652,
|
| 848 |
"step": 2320
|
| 849 |
},
|
| 850 |
{
|
| 851 |
+
"loss": 1.9814,
|
| 852 |
+
"grad_norm": 4.198029041290283,
|
| 853 |
+
"learning_rate": 0.00032946859903381644,
|
| 854 |
+
"epoch": 4.069565217391304,
|
| 855 |
"step": 2340
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"loss": 1.9505,
|
| 859 |
+
"grad_norm": 3.960844039916992,
|
| 860 |
+
"learning_rate": 0.000327536231884058,
|
| 861 |
+
"epoch": 4.104347826086957,
|
| 862 |
"step": 2360
|
| 863 |
},
|
| 864 |
{
|
| 865 |
+
"loss": 1.8815,
|
| 866 |
+
"grad_norm": 4.0190300941467285,
|
| 867 |
+
"learning_rate": 0.0003256038647342995,
|
| 868 |
+
"epoch": 4.139130434782609,
|
| 869 |
"step": 2380
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"loss": 1.8365,
|
| 873 |
+
"grad_norm": 4.040708541870117,
|
| 874 |
+
"learning_rate": 0.00032367149758454105,
|
| 875 |
+
"epoch": 4.173913043478261,
|
| 876 |
"step": 2400
|
| 877 |
},
|
| 878 |
{
|
| 879 |
+
"loss": 1.84,
|
| 880 |
+
"grad_norm": 4.077364444732666,
|
| 881 |
+
"learning_rate": 0.0003217391304347826,
|
| 882 |
+
"epoch": 4.208695652173913,
|
| 883 |
"step": 2420
|
| 884 |
},
|
| 885 |
{
|
| 886 |
+
"loss": 1.8864,
|
| 887 |
+
"grad_norm": 4.267309188842773,
|
| 888 |
+
"learning_rate": 0.0003199033816425121,
|
| 889 |
+
"epoch": 4.243478260869566,
|
| 890 |
"step": 2440
|
| 891 |
},
|
| 892 |
{
|
| 893 |
+
"loss": 1.9015,
|
| 894 |
+
"grad_norm": 3.978663921356201,
|
| 895 |
+
"learning_rate": 0.00031797101449275363,
|
| 896 |
+
"epoch": 4.278260869565218,
|
| 897 |
"step": 2460
|
| 898 |
},
|
| 899 |
{
|
| 900 |
+
"loss": 1.8388,
|
| 901 |
+
"grad_norm": 4.089256763458252,
|
| 902 |
+
"learning_rate": 0.0003160386473429952,
|
| 903 |
+
"epoch": 4.3130434782608695,
|
| 904 |
"step": 2480
|
| 905 |
},
|
| 906 |
{
|
| 907 |
+
"loss": 1.7845,
|
| 908 |
+
"grad_norm": 3.9317057132720947,
|
| 909 |
+
"learning_rate": 0.0003141062801932367,
|
| 910 |
+
"epoch": 4.3478260869565215,
|
| 911 |
"step": 2500
|
| 912 |
},
|
| 913 |
{
|
| 914 |
+
"loss": 1.7725,
|
| 915 |
+
"grad_norm": 3.9738080501556396,
|
| 916 |
+
"learning_rate": 0.00031217391304347823,
|
| 917 |
+
"epoch": 4.3826086956521735,
|
| 918 |
"step": 2520
|
| 919 |
},
|
| 920 |
{
|
| 921 |
+
"loss": 1.852,
|
| 922 |
+
"grad_norm": 4.232215881347656,
|
| 923 |
+
"learning_rate": 0.0003102415458937198,
|
| 924 |
+
"epoch": 4.417391304347826,
|
| 925 |
"step": 2540
|
| 926 |
},
|
| 927 |
{
|
| 928 |
+
"loss": 1.8234,
|
| 929 |
+
"grad_norm": 4.050131797790527,
|
| 930 |
+
"learning_rate": 0.00030830917874396136,
|
| 931 |
+
"epoch": 4.452173913043478,
|
| 932 |
"step": 2560
|
| 933 |
},
|
| 934 |
{
|
| 935 |
+
"loss": 1.8148,
|
| 936 |
+
"grad_norm": 4.217935085296631,
|
| 937 |
+
"learning_rate": 0.0003063768115942029,
|
| 938 |
+
"epoch": 4.48695652173913,
|
| 939 |
"step": 2580
|
| 940 |
},
|
| 941 |
{
|
| 942 |
+
"loss": 1.7134,
|
| 943 |
+
"grad_norm": 3.9807074069976807,
|
| 944 |
+
"learning_rate": 0.0003044444444444445,
|
| 945 |
+
"epoch": 4.521739130434782,
|
| 946 |
"step": 2600
|
| 947 |
},
|
| 948 |
{
|
| 949 |
+
"loss": 1.6752,
|
| 950 |
+
"grad_norm": 4.05940580368042,
|
| 951 |
+
"learning_rate": 0.000302512077294686,
|
| 952 |
+
"epoch": 4.556521739130435,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 953 |
"step": 2620
|
| 954 |
},
|
| 955 |
{
|
| 956 |
+
"loss": 1.8413,
|
| 957 |
+
"grad_norm": 4.454566955566406,
|
| 958 |
+
"learning_rate": 0.00030057971014492755,
|
| 959 |
+
"epoch": 4.591304347826087,
|
| 960 |
"step": 2640
|
| 961 |
},
|
| 962 |
{
|
| 963 |
+
"loss": 1.7948,
|
| 964 |
+
"grad_norm": 4.144088268280029,
|
| 965 |
+
"learning_rate": 0.0002986473429951691,
|
| 966 |
+
"epoch": 4.626086956521739,
|
| 967 |
"step": 2660
|
| 968 |
},
|
| 969 |
{
|
| 970 |
+
"loss": 1.7468,
|
| 971 |
+
"grad_norm": 3.940176010131836,
|
| 972 |
+
"learning_rate": 0.0002967149758454106,
|
| 973 |
+
"epoch": 4.660869565217391,
|
| 974 |
"step": 2680
|
| 975 |
},
|
| 976 |
{
|
| 977 |
+
"loss": 1.709,
|
| 978 |
+
"grad_norm": 4.198675632476807,
|
| 979 |
+
"learning_rate": 0.0002948792270531401,
|
| 980 |
+
"epoch": 4.695652173913043,
|
| 981 |
"step": 2700
|
| 982 |
},
|
| 983 |
{
|
| 984 |
+
"loss": 1.6506,
|
| 985 |
+
"grad_norm": 3.976001501083374,
|
| 986 |
+
"learning_rate": 0.00029294685990338167,
|
| 987 |
+
"epoch": 4.730434782608696,
|
| 988 |
"step": 2720
|
| 989 |
},
|
| 990 |
{
|
| 991 |
+
"loss": 1.7042,
|
| 992 |
+
"grad_norm": 4.033059120178223,
|
| 993 |
+
"learning_rate": 0.0002910144927536232,
|
| 994 |
+
"epoch": 4.765217391304348,
|
| 995 |
"step": 2740
|
| 996 |
},
|
| 997 |
{
|
| 998 |
+
"loss": 1.6795,
|
| 999 |
+
"grad_norm": 4.062041759490967,
|
| 1000 |
+
"learning_rate": 0.0002890821256038648,
|
| 1001 |
+
"epoch": 4.8,
|
| 1002 |
"step": 2760
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
+
"loss": 1.7029,
|
| 1006 |
+
"grad_norm": 3.988589286804199,
|
| 1007 |
+
"learning_rate": 0.00028714975845410627,
|
| 1008 |
+
"epoch": 4.834782608695652,
|
| 1009 |
"step": 2780
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
+
"loss": 1.6641,
|
| 1013 |
+
"grad_norm": 4.16325044631958,
|
| 1014 |
+
"learning_rate": 0.0002852173913043478,
|
| 1015 |
+
"epoch": 4.869565217391305,
|
| 1016 |
"step": 2800
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
+
"loss": 1.6953,
|
| 1020 |
+
"grad_norm": 4.323537349700928,
|
| 1021 |
+
"learning_rate": 0.0002832850241545894,
|
| 1022 |
+
"epoch": 4.904347826086957,
|
| 1023 |
"step": 2820
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
+
"loss": 1.5863,
|
| 1027 |
+
"grad_norm": 3.8293144702911377,
|
| 1028 |
+
"learning_rate": 0.0002813526570048309,
|
| 1029 |
+
"epoch": 4.939130434782609,
|
| 1030 |
"step": 2840
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
+
"loss": 1.6276,
|
| 1034 |
+
"grad_norm": 3.8955535888671875,
|
| 1035 |
+
"learning_rate": 0.00027942028985507246,
|
| 1036 |
+
"epoch": 4.973913043478261,
|
| 1037 |
"step": 2860
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
+
"eval_loss": 0.8229038715362549,
|
| 1041 |
+
"eval_accuracy": 0.9677202420981843,
|
| 1042 |
+
"eval_runtime": 88.6744,
|
| 1043 |
+
"eval_samples_per_second": 16.769,
|
| 1044 |
+
"eval_steps_per_second": 16.769,
|
| 1045 |
+
"epoch": 5.0,
|
| 1046 |
+
"step": 2875
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"loss": 1.5701,
|
| 1050 |
+
"grad_norm": 3.8480091094970703,
|
| 1051 |
+
"learning_rate": 0.00027748792270531405,
|
| 1052 |
+
"epoch": 5.008695652173913,
|
| 1053 |
"step": 2880
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
+
"loss": 1.3786,
|
| 1057 |
+
"grad_norm": 3.679872512817383,
|
| 1058 |
+
"learning_rate": 0.0002755555555555556,
|
| 1059 |
+
"epoch": 5.043478260869565,
|
| 1060 |
"step": 2900
|
| 1061 |
},
|
| 1062 |
{
|
| 1063 |
+
"loss": 1.3563,
|
| 1064 |
+
"grad_norm": 4.13381290435791,
|
| 1065 |
+
"learning_rate": 0.00027362318840579706,
|
| 1066 |
+
"epoch": 5.078260869565217,
|
| 1067 |
"step": 2920
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
+
"loss": 1.3588,
|
| 1071 |
+
"grad_norm": 3.7467329502105713,
|
| 1072 |
+
"learning_rate": 0.00027169082125603865,
|
| 1073 |
+
"epoch": 5.113043478260869,
|
| 1074 |
"step": 2940
|
| 1075 |
},
|
| 1076 |
{
|
| 1077 |
+
"loss": 1.3782,
|
| 1078 |
+
"grad_norm": 3.5837419033050537,
|
| 1079 |
+
"learning_rate": 0.0002698550724637681,
|
| 1080 |
+
"epoch": 5.147826086956521,
|
| 1081 |
"step": 2960
|
| 1082 |
},
|
| 1083 |
{
|
| 1084 |
+
"loss": 1.3969,
|
| 1085 |
+
"grad_norm": 4.077097415924072,
|
| 1086 |
+
"learning_rate": 0.00026792270531400964,
|
| 1087 |
+
"epoch": 5.182608695652174,
|
| 1088 |
"step": 2980
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
+
"loss": 1.3346,
|
| 1092 |
+
"grad_norm": 3.5995211601257324,
|
| 1093 |
+
"learning_rate": 0.00026599033816425123,
|
| 1094 |
+
"epoch": 5.217391304347826,
|
| 1095 |
"step": 3000
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
+
"loss": 1.3772,
|
| 1099 |
+
"grad_norm": 3.714010000228882,
|
| 1100 |
+
"learning_rate": 0.00026405797101449277,
|
| 1101 |
+
"epoch": 5.252173913043478,
|
| 1102 |
"step": 3020
|
| 1103 |
},
|
| 1104 |
{
|
| 1105 |
+
"loss": 1.3452,
|
| 1106 |
+
"grad_norm": 3.807094097137451,
|
| 1107 |
+
"learning_rate": 0.00026231884057971016,
|
| 1108 |
+
"epoch": 5.28695652173913,
|
| 1109 |
"step": 3040
|
| 1110 |
},
|
| 1111 |
{
|
| 1112 |
+
"loss": 1.3161,
|
| 1113 |
+
"grad_norm": 4.012477397918701,
|
| 1114 |
+
"learning_rate": 0.0002603864734299517,
|
| 1115 |
+
"epoch": 5.321739130434783,
|
| 1116 |
"step": 3060
|
| 1117 |
},
|
| 1118 |
{
|
| 1119 |
+
"loss": 1.3146,
|
| 1120 |
+
"grad_norm": 3.850520372390747,
|
| 1121 |
+
"learning_rate": 0.0002584541062801932,
|
| 1122 |
+
"epoch": 5.356521739130435,
|
| 1123 |
"step": 3080
|
| 1124 |
},
|
| 1125 |
{
|
| 1126 |
+
"loss": 1.3057,
|
| 1127 |
+
"grad_norm": NaN,
|
| 1128 |
+
"learning_rate": 0.00025661835748792274,
|
| 1129 |
+
"epoch": 5.391304347826087,
|
| 1130 |
"step": 3100
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
+
"loss": 1.2619,
|
| 1134 |
+
"grad_norm": 3.697744607925415,
|
| 1135 |
+
"learning_rate": 0.0002546859903381643,
|
| 1136 |
+
"epoch": 5.426086956521739,
|
| 1137 |
"step": 3120
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
+
"loss": 1.3436,
|
| 1141 |
+
"grad_norm": 4.125018119812012,
|
| 1142 |
+
"learning_rate": 0.00025275362318840576,
|
| 1143 |
+
"epoch": 5.460869565217392,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1144 |
"step": 3140
|
| 1145 |
},
|
| 1146 |
{
|
| 1147 |
+
"loss": 1.3289,
|
| 1148 |
+
"grad_norm": 4.1491899490356445,
|
| 1149 |
+
"learning_rate": 0.00025082125603864735,
|
| 1150 |
+
"epoch": 5.495652173913044,
|
| 1151 |
"step": 3160
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
+
"loss": 1.218,
|
| 1155 |
+
"grad_norm": 3.9294846057891846,
|
| 1156 |
+
"learning_rate": 0.0002488888888888889,
|
| 1157 |
+
"epoch": 5.530434782608696,
|
| 1158 |
"step": 3180
|
| 1159 |
},
|
| 1160 |
{
|
| 1161 |
+
"loss": 1.3219,
|
| 1162 |
+
"grad_norm": 3.9030706882476807,
|
| 1163 |
+
"learning_rate": 0.00024695652173913047,
|
| 1164 |
+
"epoch": 5.565217391304348,
|
| 1165 |
"step": 3200
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
+
"loss": 1.2694,
|
| 1169 |
+
"grad_norm": 4.124849319458008,
|
| 1170 |
+
"learning_rate": 0.000245024154589372,
|
| 1171 |
+
"epoch": 5.6,
|
| 1172 |
"step": 3220
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
+
"loss": 1.2379,
|
| 1176 |
+
"grad_norm": 4.1668500900268555,
|
| 1177 |
+
"learning_rate": 0.0002432850241545894,
|
| 1178 |
+
"epoch": 5.6347826086956525,
|
| 1179 |
"step": 3240
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
+
"loss": 1.2892,
|
| 1183 |
+
"grad_norm": 4.098198890686035,
|
| 1184 |
+
"learning_rate": 0.00024135265700483093,
|
| 1185 |
+
"epoch": 5.6695652173913045,
|
| 1186 |
"step": 3260
|
| 1187 |
},
|
| 1188 |
{
|
| 1189 |
+
"loss": 1.2742,
|
| 1190 |
+
"grad_norm": 3.690241813659668,
|
| 1191 |
+
"learning_rate": 0.00023942028985507246,
|
| 1192 |
+
"epoch": 5.7043478260869565,
|
| 1193 |
"step": 3280
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
+
"loss": 1.1755,
|
| 1197 |
+
"grad_norm": 3.978963613510132,
|
| 1198 |
+
"learning_rate": 0.00023748792270531402,
|
| 1199 |
+
"epoch": 5.739130434782608,
|
| 1200 |
"step": 3300
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
+
"loss": 1.2256,
|
| 1204 |
+
"grad_norm": 3.7397215366363525,
|
| 1205 |
+
"learning_rate": 0.00023574879227053139,
|
| 1206 |
+
"epoch": 5.773913043478261,
|
| 1207 |
"step": 3320
|
| 1208 |
},
|
| 1209 |
{
|
| 1210 |
+
"loss": 1.238,
|
| 1211 |
+
"grad_norm": 3.9201064109802246,
|
| 1212 |
+
"learning_rate": 0.00023391304347826088,
|
| 1213 |
+
"epoch": 5.808695652173913,
|
| 1214 |
"step": 3340
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
+
"loss": 1.1706,
|
| 1218 |
+
"grad_norm": 3.725389242172241,
|
| 1219 |
+
"learning_rate": 0.0002319806763285024,
|
| 1220 |
+
"epoch": 5.843478260869565,
|
| 1221 |
"step": 3360
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
+
"loss": 1.1644,
|
| 1225 |
+
"grad_norm": 3.5844123363494873,
|
| 1226 |
+
"learning_rate": 0.00023004830917874397,
|
| 1227 |
+
"epoch": 5.878260869565217,
|
| 1228 |
"step": 3380
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
+
"loss": 1.2256,
|
| 1232 |
+
"grad_norm": 3.79936146736145,
|
| 1233 |
+
"learning_rate": 0.00022821256038647343,
|
| 1234 |
+
"epoch": 5.913043478260869,
|
| 1235 |
"step": 3400
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
+
"loss": 1.2488,
|
| 1239 |
+
"grad_norm": 3.5947725772857666,
|
| 1240 |
+
"learning_rate": 0.00022628019323671497,
|
| 1241 |
+
"epoch": 5.947826086956522,
|
| 1242 |
"step": 3420
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
+
"loss": 1.1418,
|
| 1246 |
+
"grad_norm": NaN,
|
| 1247 |
+
"learning_rate": 0.00022444444444444446,
|
| 1248 |
+
"epoch": 5.982608695652174,
|
| 1249 |
"step": 3440
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
+
"eval_loss": 0.5840117335319519,
|
| 1253 |
+
"eval_accuracy": 0.9757901815736382,
|
| 1254 |
+
"eval_runtime": 97.2696,
|
| 1255 |
+
"eval_samples_per_second": 15.287,
|
| 1256 |
+
"eval_steps_per_second": 15.287,
|
| 1257 |
+
"epoch": 6.0,
|
| 1258 |
+
"step": 3450
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"loss": 1.1254,
|
| 1262 |
+
"grad_norm": 3.5959298610687256,
|
| 1263 |
+
"learning_rate": 0.00022260869565217392,
|
| 1264 |
+
"epoch": 6.017391304347826,
|
| 1265 |
"step": 3460
|
| 1266 |
},
|
| 1267 |
{
|
| 1268 |
+
"loss": 1.0343,
|
| 1269 |
+
"grad_norm": 3.9623775482177734,
|
| 1270 |
+
"learning_rate": 0.00022067632850241545,
|
| 1271 |
+
"epoch": 6.052173913043478,
|
| 1272 |
"step": 3480
|
| 1273 |
},
|
| 1274 |
{
|
| 1275 |
+
"loss": 1.0348,
|
| 1276 |
+
"grad_norm": 3.735102415084839,
|
| 1277 |
+
"learning_rate": 0.00021874396135265702,
|
| 1278 |
+
"epoch": 6.086956521739131,
|
| 1279 |
"step": 3500
|
| 1280 |
},
|
| 1281 |
{
|
| 1282 |
+
"loss": 0.9796,
|
| 1283 |
+
"grad_norm": 3.4255013465881348,
|
| 1284 |
+
"learning_rate": 0.00021681159420289855,
|
| 1285 |
+
"epoch": 6.121739130434783,
|
| 1286 |
"step": 3520
|
| 1287 |
},
|
| 1288 |
{
|
| 1289 |
+
"loss": 0.9865,
|
| 1290 |
+
"grad_norm": 3.981841564178467,
|
| 1291 |
+
"learning_rate": 0.00021497584541062804,
|
| 1292 |
+
"epoch": 6.156521739130435,
|
| 1293 |
"step": 3540
|
| 1294 |
},
|
| 1295 |
{
|
| 1296 |
+
"loss": 1.0054,
|
| 1297 |
+
"grad_norm": 3.9057116508483887,
|
| 1298 |
+
"learning_rate": 0.00021314009661835748,
|
| 1299 |
+
"epoch": 6.191304347826087,
|
| 1300 |
"step": 3560
|
| 1301 |
},
|
| 1302 |
{
|
| 1303 |
+
"loss": 1.0012,
|
| 1304 |
+
"grad_norm": 3.626560688018799,
|
| 1305 |
+
"learning_rate": 0.00021120772946859904,
|
| 1306 |
+
"epoch": 6.226086956521739,
|
| 1307 |
"step": 3580
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
+
"loss": 1.0129,
|
| 1311 |
+
"grad_norm": 3.687683582305908,
|
| 1312 |
+
"learning_rate": 0.0002093719806763285,
|
| 1313 |
+
"epoch": 6.260869565217392,
|
| 1314 |
"step": 3600
|
| 1315 |
},
|
| 1316 |
{
|
| 1317 |
+
"loss": 0.9333,
|
| 1318 |
+
"grad_norm": 3.8632826805114746,
|
| 1319 |
+
"learning_rate": 0.00020763285024154592,
|
| 1320 |
+
"epoch": 6.2956521739130435,
|
| 1321 |
"step": 3620
|
| 1322 |
},
|
| 1323 |
{
|
| 1324 |
+
"loss": 1.0259,
|
| 1325 |
+
"grad_norm": 4.089422702789307,
|
| 1326 |
+
"learning_rate": 0.0002058937198067633,
|
| 1327 |
+
"epoch": 6.3304347826086955,
|
| 1328 |
"step": 3640
|
| 1329 |
},
|
| 1330 |
{
|
| 1331 |
+
"loss": 1.0184,
|
| 1332 |
+
"grad_norm": 4.261268615722656,
|
| 1333 |
+
"learning_rate": 0.00020415458937198067,
|
| 1334 |
+
"epoch": 6.3652173913043475,
|
| 1335 |
"step": 3660
|
| 1336 |
},
|
| 1337 |
{
|
| 1338 |
+
"loss": 1.0293,
|
| 1339 |
+
"grad_norm": 2.3901586532592773,
|
| 1340 |
+
"learning_rate": 0.0002026086956521739,
|
| 1341 |
+
"epoch": 6.4,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1342 |
"step": 3680
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
+
"loss": 1.0026,
|
| 1346 |
+
"grad_norm": 2.233633518218994,
|
| 1347 |
+
"learning_rate": 0.00020067632850241546,
|
| 1348 |
+
"epoch": 6.434782608695652,
|
| 1349 |
"step": 3700
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
+
"loss": 1.0426,
|
| 1353 |
+
"grad_norm": 2.049773693084717,
|
| 1354 |
+
"learning_rate": 0.00019893719806763285,
|
| 1355 |
+
"epoch": 6.469565217391304,
|
| 1356 |
"step": 3720
|
| 1357 |
},
|
| 1358 |
{
|
| 1359 |
+
"loss": 1.0324,
|
| 1360 |
+
"grad_norm": 2.21939754486084,
|
| 1361 |
+
"learning_rate": 0.0001970048309178744,
|
| 1362 |
+
"epoch": 6.504347826086956,
|
| 1363 |
"step": 3740
|
| 1364 |
},
|
| 1365 |
{
|
| 1366 |
+
"loss": 1.0666,
|
| 1367 |
+
"grad_norm": 2.2138895988464355,
|
| 1368 |
+
"learning_rate": 0.00019516908212560387,
|
| 1369 |
+
"epoch": 6.539130434782608,
|
| 1370 |
"step": 3760
|
| 1371 |
},
|
| 1372 |
{
|
| 1373 |
+
"loss": 1.0724,
|
| 1374 |
+
"grad_norm": 1.9186855554580688,
|
| 1375 |
+
"learning_rate": 0.0001932367149758454,
|
| 1376 |
+
"epoch": 6.573913043478261,
|
| 1377 |
"step": 3780
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
+
"loss": 1.0867,
|
| 1381 |
+
"grad_norm": 1.302451729774475,
|
| 1382 |
+
"learning_rate": 0.00019159420289855073,
|
| 1383 |
+
"epoch": 6.608695652173913,
|
| 1384 |
"step": 3800
|
| 1385 |
},
|
| 1386 |
{
|
| 1387 |
+
"loss": 1.0659,
|
| 1388 |
+
"grad_norm": 1.1770459413528442,
|
| 1389 |
+
"learning_rate": 0.00018975845410628022,
|
| 1390 |
+
"epoch": 6.643478260869565,
|
| 1391 |
"step": 3820
|
| 1392 |
},
|
| 1393 |
{
|
| 1394 |
+
"loss": 1.0494,
|
| 1395 |
+
"grad_norm": 0.2651650309562683,
|
| 1396 |
+
"learning_rate": 0.0001881159420289855,
|
| 1397 |
+
"epoch": 6.678260869565217,
|
| 1398 |
"step": 3840
|
| 1399 |
},
|
| 1400 |
{
|
| 1401 |
+
"loss": 1.0464,
|
| 1402 |
+
"grad_norm": 0.0,
|
| 1403 |
+
"learning_rate": 0.0001867632850241546,
|
| 1404 |
+
"epoch": 6.71304347826087,
|
| 1405 |
"step": 3860
|
| 1406 |
},
|
| 1407 |
{
|
| 1408 |
+
"loss": 1.0457,
|
| 1409 |
+
"grad_norm": 0.0,
|
| 1410 |
+
"learning_rate": 0.000185024154589372,
|
| 1411 |
+
"epoch": 6.747826086956522,
|
| 1412 |
"step": 3880
|
| 1413 |
},
|
| 1414 |
{
|
| 1415 |
+
"loss": 0.9815,
|
| 1416 |
+
"grad_norm": 0.0,
|
| 1417 |
+
"learning_rate": 0.00018328502415458937,
|
| 1418 |
+
"epoch": 6.782608695652174,
|
| 1419 |
"step": 3900
|
| 1420 |
},
|
| 1421 |
{
|
| 1422 |
+
"loss": 1.0094,
|
| 1423 |
+
"grad_norm": 0.0,
|
| 1424 |
+
"learning_rate": 0.0001816425120772947,
|
| 1425 |
+
"epoch": 6.817391304347826,
|
| 1426 |
"step": 3920
|
| 1427 |
},
|
| 1428 |
{
|
| 1429 |
+
"loss": 1.0023,
|
| 1430 |
+
"grad_norm": NaN,
|
| 1431 |
+
"learning_rate": 0.00018028985507246377,
|
| 1432 |
+
"epoch": 6.852173913043478,
|
| 1433 |
"step": 3940
|
| 1434 |
},
|
| 1435 |
{
|
| 1436 |
+
"loss": 1.0278,
|
| 1437 |
+
"grad_norm": 0.0,
|
| 1438 |
+
"learning_rate": 0.00017893719806763288,
|
| 1439 |
+
"epoch": 6.886956521739131,
|
| 1440 |
"step": 3960
|
| 1441 |
},
|
| 1442 |
{
|
| 1443 |
+
"loss": 1.0123,
|
| 1444 |
+
"grad_norm": 0.0,
|
| 1445 |
+
"learning_rate": 0.0001771014492753623,
|
| 1446 |
+
"epoch": 6.921739130434783,
|
| 1447 |
"step": 3980
|
| 1448 |
},
|
| 1449 |
{
|
| 1450 |
+
"loss": 1.0774,
|
| 1451 |
+
"grad_norm": 0.0,
|
| 1452 |
+
"learning_rate": 0.00017565217391304346,
|
| 1453 |
+
"epoch": 6.956521739130435,
|
| 1454 |
"step": 4000
|
| 1455 |
},
|
| 1456 |
{
|
| 1457 |
+
"loss": 1.0484,
|
| 1458 |
+
"grad_norm": 0.0,
|
| 1459 |
+
"learning_rate": 0.00017391304347826088,
|
| 1460 |
+
"epoch": 6.9913043478260875,
|
| 1461 |
"step": 4020
|
| 1462 |
},
|
| 1463 |
{
|
| 1464 |
+
"eval_loss": 0.5780686736106873,
|
| 1465 |
+
"eval_accuracy": 0.9737726967047747,
|
| 1466 |
+
"eval_runtime": 118.8154,
|
| 1467 |
+
"eval_samples_per_second": 12.515,
|
| 1468 |
+
"eval_steps_per_second": 12.515,
|
| 1469 |
+
"epoch": 7.0,
|
| 1470 |
+
"step": 4025
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"loss": 0.9799,
|
| 1474 |
+
"grad_norm": 0.0,
|
| 1475 |
+
"learning_rate": 0.0001723671497584541,
|
| 1476 |
+
"epoch": 7.026086956521739,
|
| 1477 |
"step": 4040
|
| 1478 |
},
|
| 1479 |
{
|
| 1480 |
+
"loss": 0.9588,
|
| 1481 |
+
"grad_norm": 0.0,
|
| 1482 |
+
"learning_rate": 0.00017091787439613525,
|
| 1483 |
+
"epoch": 7.060869565217391,
|
| 1484 |
"step": 4060
|
| 1485 |
},
|
| 1486 |
{
|
| 1487 |
+
"loss": 0.9421,
|
| 1488 |
+
"grad_norm": NaN,
|
| 1489 |
+
"learning_rate": 0.00016966183574879226,
|
| 1490 |
+
"epoch": 7.095652173913043,
|
| 1491 |
"step": 4080
|
| 1492 |
},
|
| 1493 |
{
|
| 1494 |
+
"loss": 0.9551,
|
| 1495 |
+
"grad_norm": 0.0,
|
| 1496 |
+
"learning_rate": 0.00016782608695652175,
|
| 1497 |
+
"epoch": 7.130434782608695,
|
| 1498 |
"step": 4100
|
| 1499 |
},
|
| 1500 |
{
|
| 1501 |
+
"loss": 0.9622,
|
| 1502 |
+
"grad_norm": 0.0,
|
| 1503 |
+
"learning_rate": 0.00016618357487922704,
|
| 1504 |
+
"epoch": 7.165217391304348,
|
| 1505 |
"step": 4120
|
| 1506 |
},
|
| 1507 |
{
|
| 1508 |
+
"loss": 0.9712,
|
| 1509 |
+
"grad_norm": 0.0,
|
| 1510 |
+
"learning_rate": 0.00016444444444444446,
|
| 1511 |
+
"epoch": 7.2,
|
| 1512 |
"step": 4140
|
| 1513 |
},
|
| 1514 |
{
|
| 1515 |
+
"loss": 0.9834,
|
| 1516 |
+
"grad_norm": 0.0,
|
| 1517 |
+
"learning_rate": 0.00016299516908212561,
|
| 1518 |
+
"epoch": 7.234782608695652,
|
| 1519 |
"step": 4160
|
| 1520 |
},
|
| 1521 |
{
|
| 1522 |
+
"loss": 0.9968,
|
| 1523 |
+
"grad_norm": NaN,
|
| 1524 |
+
"learning_rate": 0.00016135265700483093,
|
| 1525 |
+
"epoch": 7.269565217391304,
|
| 1526 |
"step": 4180
|
| 1527 |
},
|
| 1528 |
{
|
| 1529 |
+
"loss": 0.956,
|
| 1530 |
+
"grad_norm": 0.0,
|
| 1531 |
+
"learning_rate": 0.00015961352657004833,
|
| 1532 |
+
"epoch": 7.304347826086957,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1533 |
"step": 4200
|
| 1534 |
},
|
| 1535 |
{
|
| 1536 |
+
"loss": 0.8981,
|
| 1537 |
+
"grad_norm": 0.0,
|
| 1538 |
+
"learning_rate": 0.00015806763285024155,
|
| 1539 |
+
"epoch": 7.339130434782609,
|
| 1540 |
"step": 4220
|
| 1541 |
},
|
| 1542 |
{
|
| 1543 |
+
"loss": 0.9515,
|
| 1544 |
+
"grad_norm": 0.0,
|
| 1545 |
+
"learning_rate": 0.00015642512077294684,
|
| 1546 |
+
"epoch": 7.373913043478261,
|
| 1547 |
"step": 4240
|
| 1548 |
},
|
| 1549 |
{
|
| 1550 |
+
"loss": 0.9535,
|
| 1551 |
+
"grad_norm": 0.0,
|
| 1552 |
+
"learning_rate": 0.0001548792270531401,
|
| 1553 |
+
"epoch": 7.408695652173913,
|
| 1554 |
"step": 4260
|
| 1555 |
},
|
| 1556 |
{
|
| 1557 |
+
"loss": 0.9646,
|
| 1558 |
+
"grad_norm": NaN,
|
| 1559 |
+
"learning_rate": 0.00015333333333333334,
|
| 1560 |
+
"epoch": 7.443478260869565,
|
| 1561 |
"step": 4280
|
| 1562 |
},
|
| 1563 |
{
|
| 1564 |
+
"loss": 0.9821,
|
| 1565 |
+
"grad_norm": 0.0,
|
| 1566 |
+
"learning_rate": 0.00015140096618357487,
|
| 1567 |
+
"epoch": 7.478260869565218,
|
| 1568 |
"step": 4300
|
| 1569 |
},
|
| 1570 |
{
|
| 1571 |
+
"loss": 0.9259,
|
| 1572 |
+
"grad_norm": 0.0,
|
| 1573 |
+
"learning_rate": 0.00015014492753623188,
|
| 1574 |
+
"epoch": 7.51304347826087,
|
| 1575 |
"step": 4320
|
| 1576 |
},
|
| 1577 |
{
|
| 1578 |
+
"loss": 0.9494,
|
| 1579 |
+
"grad_norm": 0.0,
|
| 1580 |
+
"learning_rate": 0.00014869565217391303,
|
| 1581 |
+
"epoch": 7.547826086956522,
|
| 1582 |
"step": 4340
|
| 1583 |
},
|
| 1584 |
{
|
| 1585 |
+
"loss": 0.9305,
|
| 1586 |
+
"grad_norm": 0.0,
|
| 1587 |
+
"learning_rate": 0.00014714975845410628,
|
| 1588 |
+
"epoch": 7.582608695652174,
|
| 1589 |
"step": 4360
|
| 1590 |
},
|
| 1591 |
{
|
| 1592 |
+
"loss": 0.8889,
|
| 1593 |
+
"grad_norm": 0.0,
|
| 1594 |
+
"learning_rate": 0.0001455072463768116,
|
| 1595 |
+
"epoch": 7.6173913043478265,
|
| 1596 |
"step": 4380
|
| 1597 |
},
|
| 1598 |
{
|
| 1599 |
+
"loss": 0.9524,
|
| 1600 |
+
"grad_norm": 0.0,
|
| 1601 |
+
"learning_rate": 0.00014396135265700482,
|
| 1602 |
+
"epoch": 7.6521739130434785,
|
| 1603 |
"step": 4400
|
| 1604 |
},
|
| 1605 |
{
|
| 1606 |
+
"loss": 0.9065,
|
| 1607 |
+
"grad_norm": 0.0,
|
| 1608 |
+
"learning_rate": 0.00014231884057971014,
|
| 1609 |
+
"epoch": 7.6869565217391305,
|
| 1610 |
"step": 4420
|
| 1611 |
},
|
| 1612 |
{
|
| 1613 |
+
"loss": 0.9153,
|
| 1614 |
+
"grad_norm": 0.0,
|
| 1615 |
+
"learning_rate": 0.00014048309178743963,
|
| 1616 |
+
"epoch": 7.721739130434782,
|
| 1617 |
"step": 4440
|
| 1618 |
},
|
| 1619 |
{
|
| 1620 |
+
"loss": 0.6675,
|
| 1621 |
+
"grad_norm": NaN,
|
| 1622 |
+
"learning_rate": 0.0001403864734299517,
|
| 1623 |
+
"epoch": 7.756521739130434,
|
| 1624 |
"step": 4460
|
| 1625 |
},
|
| 1626 |
{
|
| 1627 |
+
"loss": 0.0,
|
| 1628 |
+
"grad_norm": NaN,
|
| 1629 |
+
"learning_rate": 0.0001403864734299517,
|
| 1630 |
+
"epoch": 7.791304347826087,
|
| 1631 |
"step": 4480
|
| 1632 |
},
|
| 1633 |
{
|
| 1634 |
+
"loss": 0.0,
|
| 1635 |
+
"grad_norm": NaN,
|
| 1636 |
+
"learning_rate": 0.0001403864734299517,
|
| 1637 |
+
"epoch": 7.826086956521739,
|
| 1638 |
"step": 4500
|
| 1639 |
},
|
| 1640 |
{
|
| 1641 |
+
"loss": 0.0,
|
| 1642 |
+
"grad_norm": NaN,
|
| 1643 |
+
"learning_rate": 0.0001403864734299517,
|
| 1644 |
+
"epoch": 7.860869565217391,
|
| 1645 |
"step": 4520
|
| 1646 |
},
|
| 1647 |
{
|
| 1648 |
+
"loss": 0.0,
|
| 1649 |
+
"grad_norm": NaN,
|
| 1650 |
+
"learning_rate": 0.0001403864734299517,
|
| 1651 |
+
"epoch": 7.895652173913043,
|
| 1652 |
"step": 4540
|
| 1653 |
},
|
| 1654 |
{
|
| 1655 |
+
"loss": 0.0,
|
| 1656 |
+
"grad_norm": NaN,
|
| 1657 |
+
"learning_rate": 0.0001403864734299517,
|
| 1658 |
+
"epoch": 7.930434782608696,
|
| 1659 |
"step": 4560
|
| 1660 |
},
|
| 1661 |
{
|
| 1662 |
+
"loss": 0.0,
|
| 1663 |
+
"grad_norm": NaN,
|
| 1664 |
+
"learning_rate": 0.0001403864734299517,
|
| 1665 |
+
"epoch": 7.965217391304348,
|
| 1666 |
"step": 4580
|
| 1667 |
},
|
| 1668 |
{
|
| 1669 |
+
"loss": 0.0,
|
| 1670 |
+
"grad_norm": NaN,
|
| 1671 |
+
"learning_rate": 0.0001403864734299517,
|
| 1672 |
+
"epoch": 8.0,
|
| 1673 |
"step": 4600
|
| 1674 |
},
|
| 1675 |
{
|
| 1676 |
+
"eval_loss": NaN,
|
| 1677 |
+
"eval_accuracy": 0.0006724949562878278,
|
| 1678 |
+
"eval_runtime": 129.6238,
|
| 1679 |
+
"eval_samples_per_second": 11.472,
|
| 1680 |
+
"eval_steps_per_second": 11.472,
|
| 1681 |
+
"epoch": 8.0,
|
| 1682 |
+
"step": 4600
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"loss": 0.0,
|
| 1686 |
+
"grad_norm": NaN,
|
| 1687 |
+
"learning_rate": 0.0001403864734299517,
|
| 1688 |
+
"epoch": 8.034782608695652,
|
| 1689 |
"step": 4620
|
| 1690 |
},
|
| 1691 |
{
|
| 1692 |
+
"loss": 0.0,
|
| 1693 |
+
"grad_norm": NaN,
|
| 1694 |
+
"learning_rate": 0.0001403864734299517,
|
| 1695 |
+
"epoch": 8.069565217391304,
|
| 1696 |
"step": 4640
|
| 1697 |
},
|
| 1698 |
{
|
| 1699 |
+
"loss": 0.0,
|
| 1700 |
+
"grad_norm": NaN,
|
| 1701 |
+
"learning_rate": 0.0001403864734299517,
|
| 1702 |
+
"epoch": 8.104347826086956,
|
| 1703 |
"step": 4660
|
| 1704 |
},
|
| 1705 |
{
|
| 1706 |
+
"loss": 0.0,
|
| 1707 |
+
"grad_norm": NaN,
|
| 1708 |
+
"learning_rate": 0.0001403864734299517,
|
| 1709 |
+
"epoch": 8.139130434782608,
|
| 1710 |
"step": 4680
|
| 1711 |
},
|
| 1712 |
{
|
| 1713 |
+
"loss": 0.0,
|
| 1714 |
+
"grad_norm": NaN,
|
| 1715 |
+
"learning_rate": 0.0001403864734299517,
|
| 1716 |
+
"epoch": 8.173913043478262,
|
| 1717 |
"step": 4700
|
| 1718 |
},
|
| 1719 |
{
|
| 1720 |
+
"loss": 0.0,
|
| 1721 |
+
"grad_norm": NaN,
|
| 1722 |
+
"learning_rate": 0.0001403864734299517,
|
| 1723 |
+
"epoch": 8.208695652173914,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1724 |
"step": 4720
|
| 1725 |
},
|
| 1726 |
{
|
| 1727 |
+
"loss": 0.0,
|
| 1728 |
+
"grad_norm": NaN,
|
| 1729 |
+
"learning_rate": 0.0001403864734299517,
|
| 1730 |
+
"epoch": 8.243478260869566,
|
| 1731 |
"step": 4740
|
| 1732 |
},
|
| 1733 |
{
|
| 1734 |
+
"loss": 0.0,
|
| 1735 |
+
"grad_norm": NaN,
|
| 1736 |
+
"learning_rate": 0.0001403864734299517,
|
| 1737 |
+
"epoch": 8.278260869565218,
|
| 1738 |
"step": 4760
|
| 1739 |
},
|
| 1740 |
{
|
| 1741 |
+
"loss": 0.0,
|
| 1742 |
+
"grad_norm": NaN,
|
| 1743 |
+
"learning_rate": 0.0001403864734299517,
|
| 1744 |
+
"epoch": 8.31304347826087,
|
| 1745 |
"step": 4780
|
| 1746 |
},
|
| 1747 |
{
|
| 1748 |
+
"loss": 0.0,
|
| 1749 |
+
"grad_norm": NaN,
|
| 1750 |
+
"learning_rate": 0.0001403864734299517,
|
| 1751 |
+
"epoch": 8.347826086956522,
|
| 1752 |
"step": 4800
|
| 1753 |
},
|
| 1754 |
{
|
| 1755 |
+
"loss": 0.0,
|
| 1756 |
+
"grad_norm": NaN,
|
| 1757 |
+
"learning_rate": 0.0001403864734299517,
|
| 1758 |
+
"epoch": 8.382608695652173,
|
| 1759 |
"step": 4820
|
| 1760 |
},
|
| 1761 |
{
|
| 1762 |
+
"loss": 0.0,
|
| 1763 |
+
"grad_norm": NaN,
|
| 1764 |
+
"learning_rate": 0.0001403864734299517,
|
| 1765 |
+
"epoch": 8.417391304347825,
|
| 1766 |
"step": 4840
|
| 1767 |
},
|
| 1768 |
{
|
| 1769 |
+
"loss": 0.0,
|
| 1770 |
+
"grad_norm": NaN,
|
| 1771 |
+
"learning_rate": 0.0001403864734299517,
|
| 1772 |
+
"epoch": 8.452173913043477,
|
| 1773 |
"step": 4860
|
| 1774 |
},
|
| 1775 |
{
|
| 1776 |
+
"loss": 0.0,
|
| 1777 |
+
"grad_norm": NaN,
|
| 1778 |
+
"learning_rate": 0.0001403864734299517,
|
| 1779 |
+
"epoch": 8.486956521739131,
|
| 1780 |
"step": 4880
|
| 1781 |
},
|
| 1782 |
{
|
| 1783 |
+
"loss": 0.0,
|
| 1784 |
+
"grad_norm": NaN,
|
| 1785 |
+
"learning_rate": 0.0001403864734299517,
|
| 1786 |
+
"epoch": 8.521739130434783,
|
| 1787 |
"step": 4900
|
| 1788 |
},
|
| 1789 |
{
|
| 1790 |
+
"loss": 0.0,
|
| 1791 |
+
"grad_norm": NaN,
|
| 1792 |
+
"learning_rate": 0.0001403864734299517,
|
| 1793 |
+
"epoch": 8.556521739130435,
|
| 1794 |
"step": 4920
|
| 1795 |
},
|
| 1796 |
{
|
| 1797 |
+
"loss": 0.0,
|
| 1798 |
+
"grad_norm": NaN,
|
| 1799 |
+
"learning_rate": 0.0001403864734299517,
|
| 1800 |
+
"epoch": 8.591304347826087,
|
| 1801 |
"step": 4940
|
| 1802 |
},
|
| 1803 |
{
|
| 1804 |
+
"loss": 0.0,
|
| 1805 |
+
"grad_norm": NaN,
|
| 1806 |
+
"learning_rate": 0.0001403864734299517,
|
| 1807 |
+
"epoch": 8.626086956521739,
|
| 1808 |
"step": 4960
|
| 1809 |
},
|
| 1810 |
{
|
| 1811 |
+
"loss": 0.0,
|
| 1812 |
+
"grad_norm": NaN,
|
| 1813 |
+
"learning_rate": 0.0001403864734299517,
|
| 1814 |
+
"epoch": 8.660869565217391,
|
| 1815 |
"step": 4980
|
| 1816 |
},
|
| 1817 |
{
|
| 1818 |
+
"loss": 0.0,
|
| 1819 |
+
"grad_norm": NaN,
|
| 1820 |
+
"learning_rate": 0.0001403864734299517,
|
| 1821 |
+
"epoch": 8.695652173913043,
|
| 1822 |
"step": 5000
|
| 1823 |
},
|
| 1824 |
{
|
| 1825 |
+
"loss": 0.0,
|
| 1826 |
+
"grad_norm": NaN,
|
| 1827 |
+
"learning_rate": 0.0001403864734299517,
|
| 1828 |
+
"epoch": 8.730434782608695,
|
| 1829 |
"step": 5020
|
| 1830 |
},
|
| 1831 |
{
|
| 1832 |
+
"loss": 0.0,
|
| 1833 |
+
"grad_norm": NaN,
|
| 1834 |
+
"learning_rate": 0.0001403864734299517,
|
| 1835 |
+
"epoch": 8.765217391304347,
|
| 1836 |
"step": 5040
|
| 1837 |
},
|
| 1838 |
{
|
| 1839 |
+
"loss": 0.0,
|
| 1840 |
+
"grad_norm": NaN,
|
| 1841 |
+
"learning_rate": 0.0001403864734299517,
|
| 1842 |
+
"epoch": 8.8,
|
| 1843 |
"step": 5060
|
| 1844 |
},
|
| 1845 |
{
|
| 1846 |
+
"loss": 0.0,
|
| 1847 |
+
"grad_norm": NaN,
|
| 1848 |
+
"learning_rate": 0.0001403864734299517,
|
| 1849 |
+
"epoch": 8.834782608695653,
|
| 1850 |
"step": 5080
|
| 1851 |
},
|
| 1852 |
{
|
| 1853 |
+
"loss": 0.0,
|
| 1854 |
+
"grad_norm": NaN,
|
| 1855 |
+
"learning_rate": 0.0001403864734299517,
|
| 1856 |
+
"epoch": 8.869565217391305,
|
| 1857 |
"step": 5100
|
| 1858 |
},
|
| 1859 |
{
|
| 1860 |
+
"loss": 0.0,
|
| 1861 |
+
"grad_norm": NaN,
|
| 1862 |
+
"learning_rate": 0.0001403864734299517,
|
| 1863 |
+
"epoch": 8.904347826086957,
|
| 1864 |
"step": 5120
|
| 1865 |
},
|
| 1866 |
{
|
| 1867 |
+
"loss": 0.0,
|
| 1868 |
+
"grad_norm": NaN,
|
| 1869 |
+
"learning_rate": 0.0001403864734299517,
|
| 1870 |
+
"epoch": 8.939130434782609,
|
| 1871 |
"step": 5140
|
| 1872 |
},
|
| 1873 |
{
|
| 1874 |
+
"loss": 0.0,
|
| 1875 |
+
"grad_norm": NaN,
|
| 1876 |
+
"learning_rate": 0.0001403864734299517,
|
| 1877 |
+
"epoch": 8.97391304347826,
|
| 1878 |
"step": 5160
|
| 1879 |
},
|
| 1880 |
{
|
| 1881 |
+
"eval_loss": NaN,
|
| 1882 |
+
"eval_accuracy": 0.0006724949562878278,
|
| 1883 |
+
"eval_runtime": 117.1288,
|
| 1884 |
+
"eval_samples_per_second": 12.695,
|
| 1885 |
+
"eval_steps_per_second": 12.695,
|
| 1886 |
+
"epoch": 9.0,
|
| 1887 |
+
"step": 5175
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"loss": 0.0,
|
| 1891 |
+
"grad_norm": NaN,
|
| 1892 |
+
"learning_rate": 0.0001403864734299517,
|
| 1893 |
+
"epoch": 9.008695652173913,
|
| 1894 |
"step": 5180
|
| 1895 |
},
|
| 1896 |
{
|
| 1897 |
+
"loss": 0.0,
|
| 1898 |
+
"grad_norm": NaN,
|
| 1899 |
+
"learning_rate": 0.0001403864734299517,
|
| 1900 |
+
"epoch": 9.043478260869565,
|
| 1901 |
"step": 5200
|
| 1902 |
},
|
| 1903 |
{
|
| 1904 |
+
"loss": 0.0,
|
| 1905 |
+
"grad_norm": NaN,
|
| 1906 |
+
"learning_rate": 0.0001403864734299517,
|
| 1907 |
+
"epoch": 9.078260869565218,
|
| 1908 |
"step": 5220
|
| 1909 |
},
|
| 1910 |
{
|
| 1911 |
+
"loss": 0.0,
|
| 1912 |
+
"grad_norm": NaN,
|
| 1913 |
+
"learning_rate": 0.0001403864734299517,
|
| 1914 |
+
"epoch": 9.11304347826087,
|
| 1915 |
+
"step": 5240
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"loss": 0.0,
|
| 1919 |
+
"grad_norm": NaN,
|
| 1920 |
+
"learning_rate": 0.0001403864734299517,
|
| 1921 |
+
"epoch": 9.147826086956522,
|
| 1922 |
+
"step": 5260
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"loss": 0.0,
|
| 1926 |
+
"grad_norm": NaN,
|
| 1927 |
+
"learning_rate": 0.0001403864734299517,
|
| 1928 |
+
"epoch": 9.182608695652174,
|
| 1929 |
+
"step": 5280
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"loss": 0.0,
|
| 1933 |
+
"grad_norm": NaN,
|
| 1934 |
+
"learning_rate": 0.0001403864734299517,
|
| 1935 |
+
"epoch": 9.217391304347826,
|
| 1936 |
+
"step": 5300
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"loss": 0.0,
|
| 1940 |
+
"grad_norm": NaN,
|
| 1941 |
+
"learning_rate": 0.0001403864734299517,
|
| 1942 |
+
"epoch": 9.252173913043478,
|
| 1943 |
+
"step": 5320
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"loss": 0.0,
|
| 1947 |
+
"grad_norm": NaN,
|
| 1948 |
+
"learning_rate": 0.0001403864734299517,
|
| 1949 |
+
"epoch": 9.28695652173913,
|
| 1950 |
+
"step": 5340
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"loss": 0.0,
|
| 1954 |
+
"grad_norm": NaN,
|
| 1955 |
+
"learning_rate": 0.0001403864734299517,
|
| 1956 |
+
"epoch": 9.321739130434782,
|
| 1957 |
+
"step": 5360
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"loss": 0.0,
|
| 1961 |
+
"grad_norm": NaN,
|
| 1962 |
+
"learning_rate": 0.0001403864734299517,
|
| 1963 |
+
"epoch": 9.356521739130434,
|
| 1964 |
+
"step": 5380
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"loss": 0.0,
|
| 1968 |
+
"grad_norm": NaN,
|
| 1969 |
+
"learning_rate": 0.0001403864734299517,
|
| 1970 |
+
"epoch": 9.391304347826088,
|
| 1971 |
+
"step": 5400
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"loss": 0.0,
|
| 1975 |
+
"grad_norm": NaN,
|
| 1976 |
+
"learning_rate": 0.0001403864734299517,
|
| 1977 |
+
"epoch": 9.42608695652174,
|
| 1978 |
+
"step": 5420
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"loss": 0.0,
|
| 1982 |
+
"grad_norm": NaN,
|
| 1983 |
+
"learning_rate": 0.0001403864734299517,
|
| 1984 |
+
"epoch": 9.460869565217392,
|
| 1985 |
+
"step": 5440
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"loss": 0.0,
|
| 1989 |
+
"grad_norm": NaN,
|
| 1990 |
+
"learning_rate": 0.0001403864734299517,
|
| 1991 |
+
"epoch": 9.495652173913044,
|
| 1992 |
+
"step": 5460
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"loss": 0.0,
|
| 1996 |
+
"grad_norm": NaN,
|
| 1997 |
+
"learning_rate": 0.0001403864734299517,
|
| 1998 |
+
"epoch": 9.530434782608696,
|
| 1999 |
+
"step": 5480
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"loss": 0.0,
|
| 2003 |
+
"grad_norm": NaN,
|
| 2004 |
+
"learning_rate": 0.0001403864734299517,
|
| 2005 |
+
"epoch": 9.565217391304348,
|
| 2006 |
+
"step": 5500
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"loss": 0.0,
|
| 2010 |
+
"grad_norm": NaN,
|
| 2011 |
+
"learning_rate": 0.0001403864734299517,
|
| 2012 |
+
"epoch": 9.6,
|
| 2013 |
+
"step": 5520
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"loss": 0.0,
|
| 2017 |
+
"grad_norm": NaN,
|
| 2018 |
+
"learning_rate": 0.0001403864734299517,
|
| 2019 |
+
"epoch": 9.634782608695652,
|
| 2020 |
+
"step": 5540
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"loss": 0.0,
|
| 2024 |
+
"grad_norm": NaN,
|
| 2025 |
+
"learning_rate": 0.0001403864734299517,
|
| 2026 |
+
"epoch": 9.669565217391304,
|
| 2027 |
+
"step": 5560
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"loss": 0.0,
|
| 2031 |
+
"grad_norm": NaN,
|
| 2032 |
+
"learning_rate": 0.0001403864734299517,
|
| 2033 |
+
"epoch": 9.704347826086957,
|
| 2034 |
+
"step": 5580
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"loss": 0.0,
|
| 2038 |
+
"grad_norm": NaN,
|
| 2039 |
+
"learning_rate": 0.0001403864734299517,
|
| 2040 |
+
"epoch": 9.73913043478261,
|
| 2041 |
+
"step": 5600
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"loss": 0.0,
|
| 2045 |
+
"grad_norm": NaN,
|
| 2046 |
+
"learning_rate": 0.0001403864734299517,
|
| 2047 |
+
"epoch": 9.773913043478261,
|
| 2048 |
+
"step": 5620
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"loss": 0.0,
|
| 2052 |
+
"grad_norm": NaN,
|
| 2053 |
+
"learning_rate": 0.0001403864734299517,
|
| 2054 |
+
"epoch": 9.808695652173913,
|
| 2055 |
+
"step": 5640
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"loss": 0.0,
|
| 2059 |
+
"grad_norm": NaN,
|
| 2060 |
+
"learning_rate": 0.0001403864734299517,
|
| 2061 |
+
"epoch": 9.843478260869565,
|
| 2062 |
+
"step": 5660
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"loss": 0.0,
|
| 2066 |
+
"grad_norm": NaN,
|
| 2067 |
+
"learning_rate": 0.0001403864734299517,
|
| 2068 |
+
"epoch": 9.878260869565217,
|
| 2069 |
+
"step": 5680
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"loss": 0.0,
|
| 2073 |
+
"grad_norm": NaN,
|
| 2074 |
+
"learning_rate": 0.0001403864734299517,
|
| 2075 |
+
"epoch": 9.91304347826087,
|
| 2076 |
+
"step": 5700
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"loss": 0.0,
|
| 2080 |
+
"grad_norm": NaN,
|
| 2081 |
+
"learning_rate": 0.0001403864734299517,
|
| 2082 |
+
"epoch": 9.947826086956521,
|
| 2083 |
+
"step": 5720
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"loss": 0.0,
|
| 2087 |
+
"grad_norm": NaN,
|
| 2088 |
+
"learning_rate": 0.0001403864734299517,
|
| 2089 |
+
"epoch": 9.982608695652173,
|
| 2090 |
+
"step": 5740
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"eval_loss": NaN,
|
| 2094 |
+
"eval_accuracy": 0.0006724949562878278,
|
| 2095 |
+
"eval_runtime": 103.3199,
|
| 2096 |
+
"eval_samples_per_second": 14.392,
|
| 2097 |
+
"eval_steps_per_second": 14.392,
|
| 2098 |
"epoch": 10.0,
|
| 2099 |
+
"step": 5750
|
| 2100 |
},
|
| 2101 |
{
|
| 2102 |
+
"train_runtime": 59857.6179,
|
| 2103 |
+
"train_samples_per_second": 24.584,
|
| 2104 |
+
"train_steps_per_second": 0.096,
|
| 2105 |
+
"total_flos": 2.7398100529152e+18,
|
| 2106 |
+
"train_loss": 2.9414075751926587,
|
| 2107 |
"epoch": 10.0,
|
| 2108 |
+
"step": 5750
|
| 2109 |
}
|
| 2110 |
]
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"total_flos": 2.7398100529152e+18,
|
| 4 |
+
"train_loss": 2.9414075751926587,
|
| 5 |
+
"train_runtime": 59857.6179,
|
| 6 |
+
"train_samples_per_second": 24.584,
|
| 7 |
+
"train_steps_per_second": 0.096
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1942 +1,2124 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "/mnt/data4_HDD_14TB/yang/voxceleb-checkpoints/ecapa-tdnn/voxceleb1/pretrain/c512-aam-len3-bs256-
|
| 4 |
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 6.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 13.
|
| 16 |
"step": 20
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 13.
|
| 23 |
"step": 40
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 13.
|
| 30 |
"step": 60
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 80
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 100
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 12.
|
| 51 |
"step": 120
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 12.
|
| 58 |
"step": 140
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 12.
|
| 65 |
"step": 160
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 12.
|
| 72 |
"step": 180
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss":
|
| 79 |
"step": 200
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss":
|
| 86 |
"step": 220
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss":
|
| 93 |
"step": 240
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate":
|
| 99 |
-
"loss":
|
| 100 |
"step": 260
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss":
|
| 107 |
"step": 280
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss":
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 3.
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss":
|
| 121 |
"step": 320
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm": 3.
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss":
|
| 128 |
"step": 340
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm": 3.
|
| 133 |
-
"learning_rate":
|
| 134 |
-
"loss":
|
| 135 |
"step": 360
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm": 3.
|
| 140 |
-
"learning_rate":
|
| 141 |
-
"loss":
|
| 142 |
"step": 380
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm": 3.
|
| 147 |
-
"learning_rate":
|
| 148 |
-
"loss":
|
| 149 |
"step": 400
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm": 3.
|
| 154 |
-
"learning_rate":
|
| 155 |
-
"loss":
|
| 156 |
"step": 420
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm": 3.
|
| 161 |
-
"learning_rate":
|
| 162 |
-
"loss":
|
| 163 |
"step": 440
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm": 3.
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss":
|
| 170 |
"step": 460
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm": 3.
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss":
|
| 177 |
"step": 480
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm": 3.
|
| 182 |
-
"learning_rate":
|
| 183 |
-
"loss":
|
| 184 |
"step": 500
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm": 3.
|
| 189 |
-
"learning_rate":
|
| 190 |
-
"loss":
|
| 191 |
"step": 520
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch":
|
| 195 |
-
"
|
| 196 |
-
"
|
| 197 |
-
"
|
| 198 |
-
"eval_samples_per_second": 25.083,
|
| 199 |
-
"eval_steps_per_second": 25.083,
|
| 200 |
-
"step": 523
|
| 201 |
-
},
|
| 202 |
-
{
|
| 203 |
-
"epoch": 1.0325047801147227,
|
| 204 |
-
"grad_norm": 3.6838159561157227,
|
| 205 |
-
"learning_rate": 9.963883577650308e-05,
|
| 206 |
-
"loss": 11.2589,
|
| 207 |
"step": 540
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch":
|
| 211 |
-
"grad_norm": 3.
|
| 212 |
-
"learning_rate":
|
| 213 |
-
"loss":
|
| 214 |
"step": 560
|
| 215 |
},
|
| 216 |
{
|
| 217 |
-
"epoch": 1.
|
| 218 |
-
"
|
| 219 |
-
"
|
| 220 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 221 |
"step": 580
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch": 1.
|
| 225 |
-
"grad_norm": 3.
|
| 226 |
-
"learning_rate":
|
| 227 |
-
"loss":
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch": 1.
|
| 232 |
-
"grad_norm": 3.
|
| 233 |
-
"learning_rate":
|
| 234 |
-
"loss":
|
| 235 |
"step": 620
|
| 236 |
},
|
| 237 |
{
|
| 238 |
-
"epoch": 1.
|
| 239 |
-
"grad_norm": 3.
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss":
|
| 242 |
"step": 640
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch": 1.
|
| 246 |
-
"grad_norm": 3.
|
| 247 |
-
"learning_rate":
|
| 248 |
-
"loss":
|
| 249 |
"step": 660
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch": 1.
|
| 253 |
-
"grad_norm": 3.
|
| 254 |
-
"learning_rate":
|
| 255 |
-
"loss":
|
| 256 |
"step": 680
|
| 257 |
},
|
| 258 |
{
|
| 259 |
-
"epoch": 1.
|
| 260 |
-
"grad_norm": 3.
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss":
|
| 263 |
"step": 700
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch": 1.
|
| 267 |
-
"grad_norm": 3.
|
| 268 |
-
"learning_rate":
|
| 269 |
-
"loss":
|
| 270 |
"step": 720
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch": 1.
|
| 274 |
-
"grad_norm":
|
| 275 |
-
"learning_rate":
|
| 276 |
-
"loss":
|
| 277 |
"step": 740
|
| 278 |
},
|
| 279 |
{
|
| 280 |
-
"epoch": 1.
|
| 281 |
-
"grad_norm": 3.
|
| 282 |
-
"learning_rate":
|
| 283 |
-
"loss":
|
| 284 |
"step": 760
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch": 1.
|
| 288 |
-
"grad_norm": 3.
|
| 289 |
-
"learning_rate":
|
| 290 |
-
"loss":
|
| 291 |
"step": 780
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch": 1.
|
| 295 |
-
"grad_norm": 3.
|
| 296 |
-
"learning_rate":
|
| 297 |
-
"loss":
|
| 298 |
"step": 800
|
| 299 |
},
|
| 300 |
{
|
| 301 |
-
"epoch": 1.
|
| 302 |
-
"grad_norm":
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss":
|
| 305 |
"step": 820
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch": 1.
|
| 309 |
-
"grad_norm":
|
| 310 |
-
"learning_rate":
|
| 311 |
-
"loss":
|
| 312 |
"step": 840
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch": 1.
|
| 316 |
-
"grad_norm":
|
| 317 |
-
"learning_rate":
|
| 318 |
-
"loss":
|
| 319 |
"step": 860
|
| 320 |
},
|
| 321 |
{
|
| 322 |
-
"epoch": 1.
|
| 323 |
-
"grad_norm":
|
| 324 |
-
"learning_rate":
|
| 325 |
-
"loss":
|
| 326 |
"step": 880
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch": 1.
|
| 330 |
-
"grad_norm":
|
| 331 |
-
"learning_rate":
|
| 332 |
-
"loss":
|
| 333 |
"step": 900
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch": 1.
|
| 337 |
-
"grad_norm":
|
| 338 |
-
"learning_rate":
|
| 339 |
-
"loss":
|
| 340 |
"step": 920
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"epoch": 1.
|
| 344 |
-
"grad_norm":
|
| 345 |
-
"learning_rate":
|
| 346 |
-
"loss":
|
| 347 |
"step": 940
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 1.
|
| 351 |
-
"grad_norm":
|
| 352 |
-
"learning_rate":
|
| 353 |
-
"loss":
|
| 354 |
"step": 960
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 1.
|
| 358 |
-
"grad_norm":
|
| 359 |
-
"learning_rate":
|
| 360 |
-
"loss":
|
| 361 |
"step": 980
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch": 1.
|
| 365 |
-
"grad_norm":
|
| 366 |
-
"learning_rate":
|
| 367 |
-
"loss":
|
| 368 |
"step": 1000
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch": 1.
|
| 372 |
-
"grad_norm":
|
| 373 |
-
"learning_rate":
|
| 374 |
-
"loss":
|
| 375 |
"step": 1020
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch": 1.
|
| 379 |
-
"grad_norm":
|
| 380 |
-
"learning_rate":
|
| 381 |
-
"loss":
|
| 382 |
"step": 1040
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch":
|
| 386 |
-
"
|
| 387 |
-
"
|
| 388 |
-
"
|
| 389 |
-
"eval_samples_per_second": 32.226,
|
| 390 |
-
"eval_steps_per_second": 32.226,
|
| 391 |
-
"step": 1046
|
| 392 |
-
},
|
| 393 |
-
{
|
| 394 |
-
"epoch": 2.026768642447419,
|
| 395 |
-
"grad_norm": 4.428006172180176,
|
| 396 |
-
"learning_rate": 8.859145952836202e-05,
|
| 397 |
-
"loss": 9.714,
|
| 398 |
"step": 1060
|
| 399 |
},
|
| 400 |
{
|
| 401 |
-
"epoch":
|
| 402 |
-
"grad_norm":
|
| 403 |
-
"learning_rate":
|
| 404 |
-
"loss":
|
| 405 |
"step": 1080
|
| 406 |
},
|
| 407 |
{
|
| 408 |
-
"epoch":
|
| 409 |
-
"grad_norm": 4.
|
| 410 |
-
"learning_rate":
|
| 411 |
-
"loss":
|
| 412 |
"step": 1100
|
| 413 |
},
|
| 414 |
{
|
| 415 |
-
"epoch":
|
| 416 |
-
"grad_norm":
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss":
|
| 419 |
"step": 1120
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch":
|
| 423 |
-
"grad_norm": 4.
|
| 424 |
-
"learning_rate":
|
| 425 |
-
"loss":
|
| 426 |
"step": 1140
|
| 427 |
},
|
| 428 |
{
|
| 429 |
-
"epoch": 2.
|
| 430 |
-
"
|
| 431 |
-
"
|
| 432 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 433 |
"step": 1160
|
| 434 |
},
|
| 435 |
{
|
| 436 |
-
"epoch": 2.
|
| 437 |
-
"grad_norm":
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss":
|
| 440 |
"step": 1180
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"epoch": 2.
|
| 444 |
-
"grad_norm": 4.
|
| 445 |
-
"learning_rate":
|
| 446 |
-
"loss":
|
| 447 |
"step": 1200
|
| 448 |
},
|
| 449 |
{
|
| 450 |
-
"epoch": 2.
|
| 451 |
-
"grad_norm": 4.
|
| 452 |
-
"learning_rate":
|
| 453 |
-
"loss":
|
| 454 |
"step": 1220
|
| 455 |
},
|
| 456 |
{
|
| 457 |
-
"epoch": 2.
|
| 458 |
-
"grad_norm": 4.
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss":
|
| 461 |
"step": 1240
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch": 2.
|
| 465 |
-
"grad_norm": 4.
|
| 466 |
-
"learning_rate":
|
| 467 |
-
"loss":
|
| 468 |
"step": 1260
|
| 469 |
},
|
| 470 |
{
|
| 471 |
-
"epoch": 2.
|
| 472 |
-
"grad_norm": 4.
|
| 473 |
-
"learning_rate":
|
| 474 |
-
"loss":
|
| 475 |
"step": 1280
|
| 476 |
},
|
| 477 |
{
|
| 478 |
-
"epoch": 2.
|
| 479 |
-
"grad_norm":
|
| 480 |
-
"learning_rate":
|
| 481 |
-
"loss":
|
| 482 |
"step": 1300
|
| 483 |
},
|
| 484 |
{
|
| 485 |
-
"epoch": 2.
|
| 486 |
-
"grad_norm": 4.
|
| 487 |
-
"learning_rate":
|
| 488 |
-
"loss":
|
| 489 |
"step": 1320
|
| 490 |
},
|
| 491 |
{
|
| 492 |
-
"epoch": 2.
|
| 493 |
-
"grad_norm": 4.
|
| 494 |
-
"learning_rate":
|
| 495 |
-
"loss":
|
| 496 |
"step": 1340
|
| 497 |
},
|
| 498 |
{
|
| 499 |
-
"epoch": 2.
|
| 500 |
-
"grad_norm": 4.
|
| 501 |
-
"learning_rate":
|
| 502 |
-
"loss":
|
| 503 |
"step": 1360
|
| 504 |
},
|
| 505 |
{
|
| 506 |
-
"epoch": 2.
|
| 507 |
-
"grad_norm": 4.
|
| 508 |
-
"learning_rate":
|
| 509 |
-
"loss":
|
| 510 |
"step": 1380
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch": 2.
|
| 514 |
-
"grad_norm": 4.
|
| 515 |
-
"learning_rate":
|
| 516 |
-
"loss":
|
| 517 |
"step": 1400
|
| 518 |
},
|
| 519 |
{
|
| 520 |
-
"epoch": 2.
|
| 521 |
-
"grad_norm": 4.
|
| 522 |
-
"learning_rate":
|
| 523 |
-
"loss":
|
| 524 |
"step": 1420
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch": 2.
|
| 528 |
-
"grad_norm": 4.
|
| 529 |
-
"learning_rate":
|
| 530 |
-
"loss":
|
| 531 |
"step": 1440
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"epoch": 2.
|
| 535 |
-
"grad_norm": 4.
|
| 536 |
-
"learning_rate":
|
| 537 |
-
"loss":
|
| 538 |
"step": 1460
|
| 539 |
},
|
| 540 |
{
|
| 541 |
-
"epoch": 2.
|
| 542 |
-
"grad_norm": 4.
|
| 543 |
-
"learning_rate":
|
| 544 |
-
"loss":
|
| 545 |
"step": 1480
|
| 546 |
},
|
| 547 |
{
|
| 548 |
-
"epoch": 2.
|
| 549 |
-
"grad_norm": 4.
|
| 550 |
-
"learning_rate":
|
| 551 |
-
"loss":
|
| 552 |
"step": 1500
|
| 553 |
},
|
| 554 |
{
|
| 555 |
-
"epoch": 2.
|
| 556 |
-
"grad_norm": 4.
|
| 557 |
-
"learning_rate":
|
| 558 |
-
"loss":
|
| 559 |
"step": 1520
|
| 560 |
},
|
| 561 |
{
|
| 562 |
-
"epoch": 2.
|
| 563 |
-
"grad_norm":
|
| 564 |
-
"learning_rate":
|
| 565 |
-
"loss":
|
| 566 |
"step": 1540
|
| 567 |
},
|
| 568 |
{
|
| 569 |
-
"epoch": 2.
|
| 570 |
-
"grad_norm":
|
| 571 |
-
"learning_rate":
|
| 572 |
-
"loss":
|
| 573 |
"step": 1560
|
| 574 |
},
|
| 575 |
{
|
| 576 |
-
"epoch":
|
| 577 |
-
"
|
| 578 |
-
"
|
| 579 |
-
"
|
| 580 |
-
"eval_samples_per_second": 35.513,
|
| 581 |
-
"eval_steps_per_second": 35.513,
|
| 582 |
-
"step": 1569
|
| 583 |
-
},
|
| 584 |
-
{
|
| 585 |
-
"epoch": 3.0210325047801145,
|
| 586 |
-
"grad_norm": 4.895749092102051,
|
| 587 |
-
"learning_rate": 7.754408328022095e-05,
|
| 588 |
-
"loss": 8.6104,
|
| 589 |
"step": 1580
|
| 590 |
},
|
| 591 |
{
|
| 592 |
-
"epoch":
|
| 593 |
-
"grad_norm":
|
| 594 |
-
"learning_rate":
|
| 595 |
-
"loss":
|
| 596 |
"step": 1600
|
| 597 |
},
|
| 598 |
{
|
| 599 |
-
"epoch":
|
| 600 |
-
"grad_norm":
|
| 601 |
-
"learning_rate":
|
| 602 |
-
"loss":
|
| 603 |
"step": 1620
|
| 604 |
},
|
| 605 |
{
|
| 606 |
-
"epoch":
|
| 607 |
-
"grad_norm":
|
| 608 |
-
"learning_rate":
|
| 609 |
-
"loss":
|
| 610 |
"step": 1640
|
| 611 |
},
|
| 612 |
{
|
| 613 |
-
"epoch":
|
| 614 |
-
"grad_norm":
|
| 615 |
-
"learning_rate":
|
| 616 |
-
"loss":
|
| 617 |
"step": 1660
|
| 618 |
},
|
| 619 |
{
|
| 620 |
-
"epoch":
|
| 621 |
-
"grad_norm":
|
| 622 |
-
"learning_rate":
|
| 623 |
-
"loss":
|
| 624 |
"step": 1680
|
| 625 |
},
|
| 626 |
{
|
| 627 |
-
"epoch":
|
| 628 |
-
"grad_norm":
|
| 629 |
-
"learning_rate":
|
| 630 |
-
"loss":
|
| 631 |
"step": 1700
|
| 632 |
},
|
| 633 |
{
|
| 634 |
-
"epoch":
|
| 635 |
-
"grad_norm":
|
| 636 |
-
"learning_rate":
|
| 637 |
-
"loss":
|
| 638 |
"step": 1720
|
| 639 |
},
|
| 640 |
{
|
| 641 |
-
"epoch": 3.
|
| 642 |
-
"
|
| 643 |
-
"
|
| 644 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 645 |
"step": 1740
|
| 646 |
},
|
| 647 |
{
|
| 648 |
-
"epoch": 3.
|
| 649 |
-
"grad_norm":
|
| 650 |
-
"learning_rate":
|
| 651 |
-
"loss":
|
| 652 |
"step": 1760
|
| 653 |
},
|
| 654 |
{
|
| 655 |
-
"epoch": 3.
|
| 656 |
-
"grad_norm":
|
| 657 |
-
"learning_rate":
|
| 658 |
-
"loss":
|
| 659 |
"step": 1780
|
| 660 |
},
|
| 661 |
{
|
| 662 |
-
"epoch": 3.
|
| 663 |
-
"grad_norm":
|
| 664 |
-
"learning_rate":
|
| 665 |
-
"loss":
|
| 666 |
"step": 1800
|
| 667 |
},
|
| 668 |
{
|
| 669 |
-
"epoch": 3.
|
| 670 |
-
"grad_norm":
|
| 671 |
-
"learning_rate":
|
| 672 |
-
"loss":
|
| 673 |
"step": 1820
|
| 674 |
},
|
| 675 |
{
|
| 676 |
-
"epoch": 3.
|
| 677 |
-
"grad_norm":
|
| 678 |
-
"learning_rate":
|
| 679 |
-
"loss":
|
| 680 |
"step": 1840
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch": 3.
|
| 684 |
-
"grad_norm":
|
| 685 |
-
"learning_rate":
|
| 686 |
-
"loss":
|
| 687 |
"step": 1860
|
| 688 |
},
|
| 689 |
{
|
| 690 |
-
"epoch": 3.
|
| 691 |
-
"grad_norm":
|
| 692 |
-
"learning_rate":
|
| 693 |
-
"loss":
|
| 694 |
"step": 1880
|
| 695 |
},
|
| 696 |
{
|
| 697 |
-
"epoch": 3.
|
| 698 |
-
"grad_norm":
|
| 699 |
-
"learning_rate":
|
| 700 |
-
"loss":
|
| 701 |
"step": 1900
|
| 702 |
},
|
| 703 |
{
|
| 704 |
-
"epoch": 3.
|
| 705 |
-
"grad_norm":
|
| 706 |
-
"learning_rate":
|
| 707 |
-
"loss":
|
| 708 |
"step": 1920
|
| 709 |
},
|
| 710 |
{
|
| 711 |
-
"epoch": 3.
|
| 712 |
-
"grad_norm":
|
| 713 |
-
"learning_rate":
|
| 714 |
-
"loss":
|
| 715 |
"step": 1940
|
| 716 |
},
|
| 717 |
{
|
| 718 |
-
"epoch": 3.
|
| 719 |
-
"grad_norm":
|
| 720 |
-
"learning_rate":
|
| 721 |
-
"loss":
|
| 722 |
"step": 1960
|
| 723 |
},
|
| 724 |
{
|
| 725 |
-
"epoch": 3.
|
| 726 |
-
"grad_norm":
|
| 727 |
-
"learning_rate":
|
| 728 |
-
"loss":
|
| 729 |
"step": 1980
|
| 730 |
},
|
| 731 |
{
|
| 732 |
-
"epoch": 3.
|
| 733 |
-
"grad_norm":
|
| 734 |
-
"learning_rate":
|
| 735 |
-
"loss":
|
| 736 |
"step": 2000
|
| 737 |
},
|
| 738 |
{
|
| 739 |
-
"epoch": 3.
|
| 740 |
-
"grad_norm":
|
| 741 |
-
"learning_rate":
|
| 742 |
-
"loss":
|
| 743 |
"step": 2020
|
| 744 |
},
|
| 745 |
{
|
| 746 |
-
"epoch": 3.
|
| 747 |
-
"grad_norm":
|
| 748 |
-
"learning_rate":
|
| 749 |
-
"loss":
|
| 750 |
"step": 2040
|
| 751 |
},
|
| 752 |
{
|
| 753 |
-
"epoch": 3.
|
| 754 |
-
"grad_norm":
|
| 755 |
-
"learning_rate":
|
| 756 |
-
"loss":
|
| 757 |
"step": 2060
|
| 758 |
},
|
| 759 |
{
|
| 760 |
-
"epoch": 3.
|
| 761 |
-
"grad_norm":
|
| 762 |
-
"learning_rate":
|
| 763 |
-
"loss":
|
| 764 |
"step": 2080
|
| 765 |
},
|
| 766 |
{
|
| 767 |
-
"epoch":
|
| 768 |
-
"
|
| 769 |
-
"
|
| 770 |
-
"
|
| 771 |
-
"eval_samples_per_second": 40.482,
|
| 772 |
-
"eval_steps_per_second": 40.482,
|
| 773 |
-
"step": 2092
|
| 774 |
-
},
|
| 775 |
-
{
|
| 776 |
-
"epoch": 4.015296367112811,
|
| 777 |
-
"grad_norm": 5.820804595947266,
|
| 778 |
-
"learning_rate": 6.649670703207989e-05,
|
| 779 |
-
"loss": 7.8607,
|
| 780 |
"step": 2100
|
| 781 |
},
|
| 782 |
{
|
| 783 |
-
"epoch":
|
| 784 |
-
"grad_norm":
|
| 785 |
-
"learning_rate":
|
| 786 |
-
"loss":
|
| 787 |
"step": 2120
|
| 788 |
},
|
| 789 |
{
|
| 790 |
-
"epoch":
|
| 791 |
-
"grad_norm":
|
| 792 |
-
"learning_rate":
|
| 793 |
-
"loss":
|
| 794 |
"step": 2140
|
| 795 |
},
|
| 796 |
{
|
| 797 |
-
"epoch":
|
| 798 |
-
"grad_norm":
|
| 799 |
-
"learning_rate":
|
| 800 |
-
"loss":
|
| 801 |
"step": 2160
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"epoch":
|
| 805 |
-
"grad_norm":
|
| 806 |
-
"learning_rate":
|
| 807 |
-
"loss":
|
| 808 |
"step": 2180
|
| 809 |
},
|
| 810 |
{
|
| 811 |
-
"epoch":
|
| 812 |
-
"grad_norm":
|
| 813 |
-
"learning_rate":
|
| 814 |
-
"loss":
|
| 815 |
"step": 2200
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch":
|
| 819 |
-
"grad_norm":
|
| 820 |
-
"learning_rate":
|
| 821 |
-
"loss":
|
| 822 |
"step": 2220
|
| 823 |
},
|
| 824 |
{
|
| 825 |
-
"epoch":
|
| 826 |
-
"grad_norm":
|
| 827 |
-
"learning_rate":
|
| 828 |
-
"loss":
|
| 829 |
"step": 2240
|
| 830 |
},
|
| 831 |
{
|
| 832 |
-
"epoch":
|
| 833 |
-
"grad_norm":
|
| 834 |
-
"learning_rate":
|
| 835 |
-
"loss":
|
| 836 |
"step": 2260
|
| 837 |
},
|
| 838 |
{
|
| 839 |
-
"epoch":
|
| 840 |
-
"grad_norm":
|
| 841 |
-
"learning_rate":
|
| 842 |
-
"loss":
|
| 843 |
"step": 2280
|
| 844 |
},
|
| 845 |
{
|
| 846 |
-
"epoch": 4.
|
| 847 |
-
"grad_norm":
|
| 848 |
-
"learning_rate":
|
| 849 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 850 |
"step": 2300
|
| 851 |
},
|
| 852 |
{
|
| 853 |
-
"epoch": 4.
|
| 854 |
-
"grad_norm":
|
| 855 |
-
"learning_rate":
|
| 856 |
-
"loss":
|
| 857 |
"step": 2320
|
| 858 |
},
|
| 859 |
{
|
| 860 |
-
"epoch": 4.
|
| 861 |
-
"grad_norm":
|
| 862 |
-
"learning_rate":
|
| 863 |
-
"loss":
|
| 864 |
"step": 2340
|
| 865 |
},
|
| 866 |
{
|
| 867 |
-
"epoch": 4.
|
| 868 |
-
"grad_norm":
|
| 869 |
-
"learning_rate":
|
| 870 |
-
"loss":
|
| 871 |
"step": 2360
|
| 872 |
},
|
| 873 |
{
|
| 874 |
-
"epoch": 4.
|
| 875 |
-
"grad_norm":
|
| 876 |
-
"learning_rate":
|
| 877 |
-
"loss":
|
| 878 |
"step": 2380
|
| 879 |
},
|
| 880 |
{
|
| 881 |
-
"epoch": 4.
|
| 882 |
-
"grad_norm":
|
| 883 |
-
"learning_rate":
|
| 884 |
-
"loss":
|
| 885 |
"step": 2400
|
| 886 |
},
|
| 887 |
{
|
| 888 |
-
"epoch": 4.
|
| 889 |
-
"grad_norm":
|
| 890 |
-
"learning_rate":
|
| 891 |
-
"loss":
|
| 892 |
"step": 2420
|
| 893 |
},
|
| 894 |
{
|
| 895 |
-
"epoch": 4.
|
| 896 |
-
"grad_norm":
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss":
|
| 899 |
"step": 2440
|
| 900 |
},
|
| 901 |
{
|
| 902 |
-
"epoch": 4.
|
| 903 |
-
"grad_norm":
|
| 904 |
-
"learning_rate":
|
| 905 |
-
"loss":
|
| 906 |
"step": 2460
|
| 907 |
},
|
| 908 |
{
|
| 909 |
-
"epoch": 4.
|
| 910 |
-
"grad_norm":
|
| 911 |
-
"learning_rate":
|
| 912 |
-
"loss":
|
| 913 |
"step": 2480
|
| 914 |
},
|
| 915 |
{
|
| 916 |
-
"epoch": 4.
|
| 917 |
-
"grad_norm":
|
| 918 |
-
"learning_rate":
|
| 919 |
-
"loss":
|
| 920 |
"step": 2500
|
| 921 |
},
|
| 922 |
{
|
| 923 |
-
"epoch": 4.
|
| 924 |
-
"grad_norm":
|
| 925 |
-
"learning_rate":
|
| 926 |
-
"loss":
|
| 927 |
"step": 2520
|
| 928 |
},
|
| 929 |
{
|
| 930 |
-
"epoch": 4.
|
| 931 |
-
"grad_norm":
|
| 932 |
-
"learning_rate":
|
| 933 |
-
"loss":
|
| 934 |
"step": 2540
|
| 935 |
},
|
| 936 |
{
|
| 937 |
-
"epoch": 4.
|
| 938 |
-
"grad_norm":
|
| 939 |
-
"learning_rate":
|
| 940 |
-
"loss":
|
| 941 |
"step": 2560
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"epoch": 4.
|
| 945 |
-
"grad_norm":
|
| 946 |
-
"learning_rate":
|
| 947 |
-
"loss":
|
| 948 |
"step": 2580
|
| 949 |
},
|
| 950 |
{
|
| 951 |
-
"epoch": 4.
|
| 952 |
-
"grad_norm":
|
| 953 |
-
"learning_rate":
|
| 954 |
-
"loss":
|
| 955 |
"step": 2600
|
| 956 |
},
|
| 957 |
{
|
| 958 |
-
"epoch":
|
| 959 |
-
"
|
| 960 |
-
"
|
| 961 |
-
"
|
| 962 |
-
"eval_samples_per_second": 194.433,
|
| 963 |
-
"eval_steps_per_second": 194.433,
|
| 964 |
-
"step": 2615
|
| 965 |
-
},
|
| 966 |
-
{
|
| 967 |
-
"epoch": 5.009560229445507,
|
| 968 |
-
"grad_norm": 6.838705062866211,
|
| 969 |
-
"learning_rate": 5.544933078393881e-05,
|
| 970 |
-
"loss": 7.1863,
|
| 971 |
"step": 2620
|
| 972 |
},
|
| 973 |
{
|
| 974 |
-
"epoch":
|
| 975 |
-
"grad_norm":
|
| 976 |
-
"learning_rate":
|
| 977 |
-
"loss":
|
| 978 |
"step": 2640
|
| 979 |
},
|
| 980 |
{
|
| 981 |
-
"epoch":
|
| 982 |
-
"grad_norm":
|
| 983 |
-
"learning_rate":
|
| 984 |
-
"loss":
|
| 985 |
"step": 2660
|
| 986 |
},
|
| 987 |
{
|
| 988 |
-
"epoch":
|
| 989 |
-
"grad_norm":
|
| 990 |
-
"learning_rate":
|
| 991 |
-
"loss":
|
| 992 |
"step": 2680
|
| 993 |
},
|
| 994 |
{
|
| 995 |
-
"epoch":
|
| 996 |
-
"grad_norm":
|
| 997 |
-
"learning_rate":
|
| 998 |
-
"loss":
|
| 999 |
"step": 2700
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
-
"epoch":
|
| 1003 |
-
"grad_norm":
|
| 1004 |
-
"learning_rate":
|
| 1005 |
-
"loss":
|
| 1006 |
"step": 2720
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
-
"epoch":
|
| 1010 |
-
"grad_norm":
|
| 1011 |
-
"learning_rate":
|
| 1012 |
-
"loss":
|
| 1013 |
"step": 2740
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
-
"epoch":
|
| 1017 |
-
"grad_norm":
|
| 1018 |
-
"learning_rate":
|
| 1019 |
-
"loss":
|
| 1020 |
"step": 2760
|
| 1021 |
},
|
| 1022 |
{
|
| 1023 |
-
"epoch":
|
| 1024 |
-
"grad_norm":
|
| 1025 |
-
"learning_rate":
|
| 1026 |
-
"loss":
|
| 1027 |
"step": 2780
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
-
"epoch":
|
| 1031 |
-
"grad_norm":
|
| 1032 |
-
"learning_rate":
|
| 1033 |
-
"loss":
|
| 1034 |
"step": 2800
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
-
"epoch":
|
| 1038 |
-
"grad_norm":
|
| 1039 |
-
"learning_rate":
|
| 1040 |
-
"loss":
|
| 1041 |
"step": 2820
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
-
"epoch":
|
| 1045 |
-
"grad_norm":
|
| 1046 |
-
"learning_rate":
|
| 1047 |
-
"loss":
|
| 1048 |
"step": 2840
|
| 1049 |
},
|
| 1050 |
{
|
| 1051 |
-
"epoch":
|
| 1052 |
-
"grad_norm":
|
| 1053 |
-
"learning_rate":
|
| 1054 |
-
"loss":
|
| 1055 |
"step": 2860
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
-
"epoch": 5.
|
| 1059 |
-
"
|
| 1060 |
-
"
|
| 1061 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1062 |
"step": 2880
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
-
"epoch": 5.
|
| 1066 |
-
"grad_norm":
|
| 1067 |
-
"learning_rate":
|
| 1068 |
-
"loss":
|
| 1069 |
"step": 2900
|
| 1070 |
},
|
| 1071 |
{
|
| 1072 |
-
"epoch": 5.
|
| 1073 |
-
"grad_norm":
|
| 1074 |
-
"learning_rate":
|
| 1075 |
-
"loss":
|
| 1076 |
"step": 2920
|
| 1077 |
},
|
| 1078 |
{
|
| 1079 |
-
"epoch": 5.
|
| 1080 |
-
"grad_norm":
|
| 1081 |
-
"learning_rate":
|
| 1082 |
-
"loss":
|
| 1083 |
"step": 2940
|
| 1084 |
},
|
| 1085 |
{
|
| 1086 |
-
"epoch": 5.
|
| 1087 |
-
"grad_norm":
|
| 1088 |
-
"learning_rate":
|
| 1089 |
-
"loss":
|
| 1090 |
"step": 2960
|
| 1091 |
},
|
| 1092 |
{
|
| 1093 |
-
"epoch": 5.
|
| 1094 |
-
"grad_norm":
|
| 1095 |
-
"learning_rate":
|
| 1096 |
-
"loss":
|
| 1097 |
"step": 2980
|
| 1098 |
},
|
| 1099 |
{
|
| 1100 |
-
"epoch": 5.
|
| 1101 |
-
"grad_norm":
|
| 1102 |
-
"learning_rate":
|
| 1103 |
-
"loss":
|
| 1104 |
"step": 3000
|
| 1105 |
},
|
| 1106 |
{
|
| 1107 |
-
"epoch": 5.
|
| 1108 |
-
"grad_norm":
|
| 1109 |
-
"learning_rate":
|
| 1110 |
-
"loss":
|
| 1111 |
"step": 3020
|
| 1112 |
},
|
| 1113 |
{
|
| 1114 |
-
"epoch": 5.
|
| 1115 |
-
"grad_norm":
|
| 1116 |
-
"learning_rate":
|
| 1117 |
-
"loss":
|
| 1118 |
"step": 3040
|
| 1119 |
},
|
| 1120 |
{
|
| 1121 |
-
"epoch": 5.
|
| 1122 |
-
"grad_norm":
|
| 1123 |
-
"learning_rate":
|
| 1124 |
-
"loss":
|
| 1125 |
"step": 3060
|
| 1126 |
},
|
| 1127 |
{
|
| 1128 |
-
"epoch": 5.
|
| 1129 |
-
"grad_norm":
|
| 1130 |
-
"learning_rate":
|
| 1131 |
-
"loss":
|
| 1132 |
"step": 3080
|
| 1133 |
},
|
| 1134 |
{
|
| 1135 |
-
"epoch": 5.
|
| 1136 |
-
"grad_norm":
|
| 1137 |
-
"learning_rate":
|
| 1138 |
-
"loss":
|
| 1139 |
"step": 3100
|
| 1140 |
},
|
| 1141 |
{
|
| 1142 |
-
"epoch": 5.
|
| 1143 |
-
"grad_norm":
|
| 1144 |
-
"learning_rate":
|
| 1145 |
-
"loss":
|
| 1146 |
"step": 3120
|
| 1147 |
},
|
| 1148 |
{
|
| 1149 |
-
"epoch":
|
| 1150 |
-
"
|
| 1151 |
-
"
|
| 1152 |
-
"
|
| 1153 |
-
"eval_samples_per_second": 33.48,
|
| 1154 |
-
"eval_steps_per_second": 33.48,
|
| 1155 |
-
"step": 3138
|
| 1156 |
-
},
|
| 1157 |
-
{
|
| 1158 |
-
"epoch": 6.003824091778203,
|
| 1159 |
-
"grad_norm": 7.1014723777771,
|
| 1160 |
-
"learning_rate": 4.440195453579775e-05,
|
| 1161 |
-
"loss": 6.6927,
|
| 1162 |
"step": 3140
|
| 1163 |
},
|
| 1164 |
{
|
| 1165 |
-
"epoch":
|
| 1166 |
-
"grad_norm":
|
| 1167 |
-
"learning_rate":
|
| 1168 |
-
"loss":
|
| 1169 |
"step": 3160
|
| 1170 |
},
|
| 1171 |
{
|
| 1172 |
-
"epoch":
|
| 1173 |
-
"grad_norm":
|
| 1174 |
-
"learning_rate":
|
| 1175 |
-
"loss":
|
| 1176 |
"step": 3180
|
| 1177 |
},
|
| 1178 |
{
|
| 1179 |
-
"epoch":
|
| 1180 |
-
"grad_norm":
|
| 1181 |
-
"learning_rate":
|
| 1182 |
-
"loss":
|
| 1183 |
"step": 3200
|
| 1184 |
},
|
| 1185 |
{
|
| 1186 |
-
"epoch": 6
|
| 1187 |
-
"grad_norm":
|
| 1188 |
-
"learning_rate":
|
| 1189 |
-
"loss":
|
| 1190 |
"step": 3220
|
| 1191 |
},
|
| 1192 |
{
|
| 1193 |
-
"epoch":
|
| 1194 |
-
"grad_norm":
|
| 1195 |
-
"learning_rate":
|
| 1196 |
-
"loss":
|
| 1197 |
"step": 3240
|
| 1198 |
},
|
| 1199 |
{
|
| 1200 |
-
"epoch":
|
| 1201 |
-
"grad_norm":
|
| 1202 |
-
"learning_rate":
|
| 1203 |
-
"loss":
|
| 1204 |
"step": 3260
|
| 1205 |
},
|
| 1206 |
{
|
| 1207 |
-
"epoch":
|
| 1208 |
-
"grad_norm":
|
| 1209 |
-
"learning_rate":
|
| 1210 |
-
"loss":
|
| 1211 |
"step": 3280
|
| 1212 |
},
|
| 1213 |
{
|
| 1214 |
-
"epoch":
|
| 1215 |
-
"grad_norm":
|
| 1216 |
-
"learning_rate":
|
| 1217 |
-
"loss":
|
| 1218 |
"step": 3300
|
| 1219 |
},
|
| 1220 |
{
|
| 1221 |
-
"epoch":
|
| 1222 |
-
"grad_norm":
|
| 1223 |
-
"learning_rate":
|
| 1224 |
-
"loss":
|
| 1225 |
"step": 3320
|
| 1226 |
},
|
| 1227 |
{
|
| 1228 |
-
"epoch":
|
| 1229 |
-
"grad_norm":
|
| 1230 |
-
"learning_rate":
|
| 1231 |
-
"loss":
|
| 1232 |
"step": 3340
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
-
"epoch":
|
| 1236 |
-
"grad_norm":
|
| 1237 |
-
"learning_rate":
|
| 1238 |
-
"loss":
|
| 1239 |
"step": 3360
|
| 1240 |
},
|
| 1241 |
{
|
| 1242 |
-
"epoch":
|
| 1243 |
-
"grad_norm":
|
| 1244 |
-
"learning_rate":
|
| 1245 |
-
"loss":
|
| 1246 |
"step": 3380
|
| 1247 |
},
|
| 1248 |
{
|
| 1249 |
-
"epoch":
|
| 1250 |
-
"grad_norm":
|
| 1251 |
-
"learning_rate":
|
| 1252 |
-
"loss":
|
| 1253 |
"step": 3400
|
| 1254 |
},
|
| 1255 |
{
|
| 1256 |
-
"epoch":
|
| 1257 |
-
"grad_norm":
|
| 1258 |
-
"learning_rate":
|
| 1259 |
-
"loss":
|
| 1260 |
"step": 3420
|
| 1261 |
},
|
| 1262 |
{
|
| 1263 |
-
"epoch":
|
| 1264 |
-
"grad_norm":
|
| 1265 |
-
"learning_rate":
|
| 1266 |
-
"loss":
|
| 1267 |
"step": 3440
|
| 1268 |
},
|
| 1269 |
{
|
| 1270 |
-
"epoch": 6.
|
| 1271 |
-
"
|
| 1272 |
-
"
|
| 1273 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1274 |
"step": 3460
|
| 1275 |
},
|
| 1276 |
{
|
| 1277 |
-
"epoch": 6.
|
| 1278 |
-
"grad_norm":
|
| 1279 |
-
"learning_rate":
|
| 1280 |
-
"loss":
|
| 1281 |
"step": 3480
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
-
"epoch": 6.
|
| 1285 |
-
"grad_norm":
|
| 1286 |
-
"learning_rate":
|
| 1287 |
-
"loss":
|
| 1288 |
"step": 3500
|
| 1289 |
},
|
| 1290 |
{
|
| 1291 |
-
"epoch": 6.
|
| 1292 |
-
"grad_norm":
|
| 1293 |
-
"learning_rate":
|
| 1294 |
-
"loss":
|
| 1295 |
"step": 3520
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
-
"epoch": 6.
|
| 1299 |
-
"grad_norm":
|
| 1300 |
-
"learning_rate":
|
| 1301 |
-
"loss":
|
| 1302 |
"step": 3540
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
-
"epoch": 6.
|
| 1306 |
-
"grad_norm":
|
| 1307 |
-
"learning_rate":
|
| 1308 |
-
"loss":
|
| 1309 |
"step": 3560
|
| 1310 |
},
|
| 1311 |
{
|
| 1312 |
-
"epoch": 6.
|
| 1313 |
-
"grad_norm":
|
| 1314 |
-
"learning_rate":
|
| 1315 |
-
"loss":
|
| 1316 |
"step": 3580
|
| 1317 |
},
|
| 1318 |
{
|
| 1319 |
-
"epoch": 6.
|
| 1320 |
-
"grad_norm":
|
| 1321 |
-
"learning_rate":
|
| 1322 |
-
"loss":
|
| 1323 |
"step": 3600
|
| 1324 |
},
|
| 1325 |
{
|
| 1326 |
-
"epoch": 6.
|
| 1327 |
-
"grad_norm":
|
| 1328 |
-
"learning_rate":
|
| 1329 |
-
"loss":
|
| 1330 |
"step": 3620
|
| 1331 |
},
|
| 1332 |
{
|
| 1333 |
-
"epoch": 6.
|
| 1334 |
-
"grad_norm":
|
| 1335 |
-
"learning_rate":
|
| 1336 |
-
"loss":
|
| 1337 |
"step": 3640
|
| 1338 |
},
|
| 1339 |
{
|
| 1340 |
-
"epoch": 6.
|
| 1341 |
-
"grad_norm":
|
| 1342 |
-
"learning_rate":
|
| 1343 |
-
"loss":
|
| 1344 |
"step": 3660
|
| 1345 |
},
|
| 1346 |
{
|
| 1347 |
-
"epoch":
|
| 1348 |
-
"
|
| 1349 |
-
"
|
| 1350 |
-
"
|
| 1351 |
-
"eval_samples_per_second": 30.688,
|
| 1352 |
-
"eval_steps_per_second": 30.688,
|
| 1353 |
-
"step": 3661
|
| 1354 |
-
},
|
| 1355 |
-
{
|
| 1356 |
-
"epoch": 7.036328871892925,
|
| 1357 |
-
"grad_norm": 7.424711227416992,
|
| 1358 |
-
"learning_rate": 3.292967920118972e-05,
|
| 1359 |
-
"loss": 6.1764,
|
| 1360 |
"step": 3680
|
| 1361 |
},
|
| 1362 |
{
|
| 1363 |
-
"epoch":
|
| 1364 |
-
"grad_norm":
|
| 1365 |
-
"learning_rate":
|
| 1366 |
-
"loss":
|
| 1367 |
"step": 3700
|
| 1368 |
},
|
| 1369 |
{
|
| 1370 |
-
"epoch":
|
| 1371 |
-
"grad_norm":
|
| 1372 |
-
"learning_rate":
|
| 1373 |
-
"loss":
|
| 1374 |
"step": 3720
|
| 1375 |
},
|
| 1376 |
{
|
| 1377 |
-
"epoch":
|
| 1378 |
-
"grad_norm":
|
| 1379 |
-
"learning_rate":
|
| 1380 |
-
"loss":
|
| 1381 |
"step": 3740
|
| 1382 |
},
|
| 1383 |
{
|
| 1384 |
-
"epoch":
|
| 1385 |
-
"grad_norm":
|
| 1386 |
-
"learning_rate":
|
| 1387 |
-
"loss":
|
| 1388 |
"step": 3760
|
| 1389 |
},
|
| 1390 |
{
|
| 1391 |
-
"epoch":
|
| 1392 |
-
"grad_norm":
|
| 1393 |
-
"learning_rate":
|
| 1394 |
-
"loss":
|
| 1395 |
"step": 3780
|
| 1396 |
},
|
| 1397 |
{
|
| 1398 |
-
"epoch":
|
| 1399 |
-
"grad_norm":
|
| 1400 |
-
"learning_rate":
|
| 1401 |
-
"loss":
|
| 1402 |
"step": 3800
|
| 1403 |
},
|
| 1404 |
{
|
| 1405 |
-
"epoch":
|
| 1406 |
-
"grad_norm":
|
| 1407 |
-
"learning_rate":
|
| 1408 |
-
"loss":
|
| 1409 |
"step": 3820
|
| 1410 |
},
|
| 1411 |
{
|
| 1412 |
-
"epoch":
|
| 1413 |
-
"grad_norm":
|
| 1414 |
-
"learning_rate":
|
| 1415 |
-
"loss":
|
| 1416 |
"step": 3840
|
| 1417 |
},
|
| 1418 |
{
|
| 1419 |
-
"epoch":
|
| 1420 |
-
"grad_norm":
|
| 1421 |
-
"learning_rate":
|
| 1422 |
-
"loss":
|
| 1423 |
"step": 3860
|
| 1424 |
},
|
| 1425 |
{
|
| 1426 |
-
"epoch":
|
| 1427 |
-
"grad_norm":
|
| 1428 |
-
"learning_rate":
|
| 1429 |
-
"loss":
|
| 1430 |
"step": 3880
|
| 1431 |
},
|
| 1432 |
{
|
| 1433 |
-
"epoch":
|
| 1434 |
-
"grad_norm":
|
| 1435 |
-
"learning_rate":
|
| 1436 |
-
"loss":
|
| 1437 |
"step": 3900
|
| 1438 |
},
|
| 1439 |
{
|
| 1440 |
-
"epoch":
|
| 1441 |
-
"grad_norm":
|
| 1442 |
-
"learning_rate":
|
| 1443 |
-
"loss":
|
| 1444 |
"step": 3920
|
| 1445 |
},
|
| 1446 |
{
|
| 1447 |
-
"epoch":
|
| 1448 |
-
"grad_norm":
|
| 1449 |
-
"learning_rate":
|
| 1450 |
-
"loss":
|
| 1451 |
"step": 3940
|
| 1452 |
},
|
| 1453 |
{
|
| 1454 |
-
"epoch":
|
| 1455 |
-
"grad_norm":
|
| 1456 |
-
"learning_rate":
|
| 1457 |
-
"loss":
|
| 1458 |
"step": 3960
|
| 1459 |
},
|
| 1460 |
{
|
| 1461 |
-
"epoch":
|
| 1462 |
-
"grad_norm":
|
| 1463 |
-
"learning_rate":
|
| 1464 |
-
"loss":
|
| 1465 |
"step": 3980
|
| 1466 |
},
|
| 1467 |
{
|
| 1468 |
-
"epoch":
|
| 1469 |
-
"grad_norm":
|
| 1470 |
-
"learning_rate":
|
| 1471 |
-
"loss":
|
| 1472 |
"step": 4000
|
| 1473 |
},
|
| 1474 |
{
|
| 1475 |
-
"epoch":
|
| 1476 |
-
"grad_norm":
|
| 1477 |
-
"learning_rate":
|
| 1478 |
-
"loss":
|
| 1479 |
"step": 4020
|
| 1480 |
},
|
| 1481 |
{
|
| 1482 |
-
"epoch": 7.
|
| 1483 |
-
"
|
| 1484 |
-
"
|
| 1485 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1486 |
"step": 4040
|
| 1487 |
},
|
| 1488 |
{
|
| 1489 |
-
"epoch": 7.
|
| 1490 |
-
"grad_norm":
|
| 1491 |
-
"learning_rate":
|
| 1492 |
-
"loss":
|
| 1493 |
"step": 4060
|
| 1494 |
},
|
| 1495 |
{
|
| 1496 |
-
"epoch": 7.
|
| 1497 |
-
"grad_norm":
|
| 1498 |
-
"learning_rate":
|
| 1499 |
-
"loss":
|
| 1500 |
"step": 4080
|
| 1501 |
},
|
| 1502 |
{
|
| 1503 |
-
"epoch": 7.
|
| 1504 |
-
"grad_norm":
|
| 1505 |
-
"learning_rate":
|
| 1506 |
-
"loss":
|
| 1507 |
"step": 4100
|
| 1508 |
},
|
| 1509 |
{
|
| 1510 |
-
"epoch": 7.
|
| 1511 |
-
"grad_norm":
|
| 1512 |
-
"learning_rate":
|
| 1513 |
-
"loss":
|
| 1514 |
"step": 4120
|
| 1515 |
},
|
| 1516 |
{
|
| 1517 |
-
"epoch": 7.
|
| 1518 |
-
"grad_norm":
|
| 1519 |
-
"learning_rate":
|
| 1520 |
-
"loss":
|
| 1521 |
"step": 4140
|
| 1522 |
},
|
| 1523 |
{
|
| 1524 |
-
"epoch": 7.
|
| 1525 |
-
"grad_norm":
|
| 1526 |
-
"learning_rate":
|
| 1527 |
-
"loss":
|
| 1528 |
"step": 4160
|
| 1529 |
},
|
| 1530 |
{
|
| 1531 |
-
"epoch": 7.
|
| 1532 |
-
"grad_norm":
|
| 1533 |
-
"learning_rate":
|
| 1534 |
-
"loss":
|
| 1535 |
"step": 4180
|
| 1536 |
},
|
| 1537 |
{
|
| 1538 |
-
"epoch":
|
| 1539 |
-
"
|
| 1540 |
-
"
|
| 1541 |
-
"
|
| 1542 |
-
"eval_samples_per_second": 57.14,
|
| 1543 |
-
"eval_steps_per_second": 57.14,
|
| 1544 |
-
"step": 4184
|
| 1545 |
-
},
|
| 1546 |
-
{
|
| 1547 |
-
"epoch": 8.030592734225621,
|
| 1548 |
-
"grad_norm": 7.686340808868408,
|
| 1549 |
-
"learning_rate": 2.1882302953048652e-05,
|
| 1550 |
-
"loss": 6.0763,
|
| 1551 |
"step": 4200
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
-
"epoch":
|
| 1555 |
-
"grad_norm":
|
| 1556 |
-
"learning_rate":
|
| 1557 |
-
"loss":
|
| 1558 |
"step": 4220
|
| 1559 |
},
|
| 1560 |
{
|
| 1561 |
-
"epoch":
|
| 1562 |
-
"grad_norm":
|
| 1563 |
-
"learning_rate":
|
| 1564 |
-
"loss":
|
| 1565 |
"step": 4240
|
| 1566 |
},
|
| 1567 |
{
|
| 1568 |
-
"epoch":
|
| 1569 |
-
"grad_norm":
|
| 1570 |
-
"learning_rate":
|
| 1571 |
-
"loss":
|
| 1572 |
"step": 4260
|
| 1573 |
},
|
| 1574 |
{
|
| 1575 |
-
"epoch":
|
| 1576 |
-
"grad_norm":
|
| 1577 |
-
"learning_rate":
|
| 1578 |
-
"loss":
|
| 1579 |
"step": 4280
|
| 1580 |
},
|
| 1581 |
{
|
| 1582 |
-
"epoch":
|
| 1583 |
-
"grad_norm":
|
| 1584 |
-
"learning_rate":
|
| 1585 |
-
"loss":
|
| 1586 |
"step": 4300
|
| 1587 |
},
|
| 1588 |
{
|
| 1589 |
-
"epoch":
|
| 1590 |
-
"grad_norm":
|
| 1591 |
-
"learning_rate":
|
| 1592 |
-
"loss":
|
| 1593 |
"step": 4320
|
| 1594 |
},
|
| 1595 |
{
|
| 1596 |
-
"epoch":
|
| 1597 |
-
"grad_norm":
|
| 1598 |
-
"learning_rate":
|
| 1599 |
-
"loss":
|
| 1600 |
"step": 4340
|
| 1601 |
},
|
| 1602 |
{
|
| 1603 |
-
"epoch":
|
| 1604 |
-
"grad_norm":
|
| 1605 |
-
"learning_rate":
|
| 1606 |
-
"loss":
|
| 1607 |
"step": 4360
|
| 1608 |
},
|
| 1609 |
{
|
| 1610 |
-
"epoch":
|
| 1611 |
-
"grad_norm":
|
| 1612 |
-
"learning_rate":
|
| 1613 |
-
"loss":
|
| 1614 |
"step": 4380
|
| 1615 |
},
|
| 1616 |
{
|
| 1617 |
-
"epoch":
|
| 1618 |
-
"grad_norm":
|
| 1619 |
-
"learning_rate":
|
| 1620 |
-
"loss":
|
| 1621 |
"step": 4400
|
| 1622 |
},
|
| 1623 |
{
|
| 1624 |
-
"epoch":
|
| 1625 |
-
"grad_norm":
|
| 1626 |
-
"learning_rate":
|
| 1627 |
-
"loss":
|
| 1628 |
"step": 4420
|
| 1629 |
},
|
| 1630 |
{
|
| 1631 |
-
"epoch":
|
| 1632 |
-
"grad_norm":
|
| 1633 |
-
"learning_rate":
|
| 1634 |
-
"loss":
|
| 1635 |
"step": 4440
|
| 1636 |
},
|
| 1637 |
{
|
| 1638 |
-
"epoch":
|
| 1639 |
-
"grad_norm":
|
| 1640 |
-
"learning_rate":
|
| 1641 |
-
"loss":
|
| 1642 |
"step": 4460
|
| 1643 |
},
|
| 1644 |
{
|
| 1645 |
-
"epoch":
|
| 1646 |
-
"grad_norm":
|
| 1647 |
-
"learning_rate":
|
| 1648 |
-
"loss":
|
| 1649 |
"step": 4480
|
| 1650 |
},
|
| 1651 |
{
|
| 1652 |
-
"epoch":
|
| 1653 |
-
"grad_norm":
|
| 1654 |
-
"learning_rate":
|
| 1655 |
-
"loss":
|
| 1656 |
"step": 4500
|
| 1657 |
},
|
| 1658 |
{
|
| 1659 |
-
"epoch":
|
| 1660 |
-
"grad_norm":
|
| 1661 |
-
"learning_rate":
|
| 1662 |
-
"loss":
|
| 1663 |
"step": 4520
|
| 1664 |
},
|
| 1665 |
{
|
| 1666 |
-
"epoch":
|
| 1667 |
-
"grad_norm":
|
| 1668 |
-
"learning_rate":
|
| 1669 |
-
"loss":
|
| 1670 |
"step": 4540
|
| 1671 |
},
|
| 1672 |
{
|
| 1673 |
-
"epoch":
|
| 1674 |
-
"grad_norm":
|
| 1675 |
-
"learning_rate":
|
| 1676 |
-
"loss":
|
| 1677 |
"step": 4560
|
| 1678 |
},
|
| 1679 |
{
|
| 1680 |
-
"epoch":
|
| 1681 |
-
"grad_norm":
|
| 1682 |
-
"learning_rate":
|
| 1683 |
-
"loss":
|
| 1684 |
"step": 4580
|
| 1685 |
},
|
| 1686 |
{
|
| 1687 |
-
"epoch": 8.
|
| 1688 |
-
"grad_norm":
|
| 1689 |
-
"learning_rate":
|
| 1690 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1691 |
"step": 4600
|
| 1692 |
},
|
| 1693 |
{
|
| 1694 |
-
"epoch": 8.
|
| 1695 |
-
"grad_norm":
|
| 1696 |
-
"learning_rate":
|
| 1697 |
-
"loss":
|
| 1698 |
"step": 4620
|
| 1699 |
},
|
| 1700 |
{
|
| 1701 |
-
"epoch": 8.
|
| 1702 |
-
"grad_norm":
|
| 1703 |
-
"learning_rate":
|
| 1704 |
-
"loss":
|
| 1705 |
"step": 4640
|
| 1706 |
},
|
| 1707 |
{
|
| 1708 |
-
"epoch": 8.
|
| 1709 |
-
"grad_norm":
|
| 1710 |
-
"learning_rate":
|
| 1711 |
-
"loss":
|
| 1712 |
"step": 4660
|
| 1713 |
},
|
| 1714 |
{
|
| 1715 |
-
"epoch": 8.
|
| 1716 |
-
"grad_norm":
|
| 1717 |
-
"learning_rate":
|
| 1718 |
-
"loss":
|
| 1719 |
"step": 4680
|
| 1720 |
},
|
| 1721 |
{
|
| 1722 |
-
"epoch": 8.
|
| 1723 |
-
"grad_norm":
|
| 1724 |
-
"learning_rate":
|
| 1725 |
-
"loss":
|
| 1726 |
"step": 4700
|
| 1727 |
},
|
| 1728 |
{
|
| 1729 |
-
"epoch":
|
| 1730 |
-
"
|
| 1731 |
-
"
|
| 1732 |
-
"
|
| 1733 |
-
"eval_samples_per_second": 27.985,
|
| 1734 |
-
"eval_steps_per_second": 27.985,
|
| 1735 |
-
"step": 4707
|
| 1736 |
-
},
|
| 1737 |
-
{
|
| 1738 |
-
"epoch": 9.024856596558317,
|
| 1739 |
-
"grad_norm": 7.77038049697876,
|
| 1740 |
-
"learning_rate": 1.0834926704907584e-05,
|
| 1741 |
-
"loss": 5.7978,
|
| 1742 |
"step": 4720
|
| 1743 |
},
|
| 1744 |
{
|
| 1745 |
-
"epoch":
|
| 1746 |
-
"grad_norm":
|
| 1747 |
-
"learning_rate":
|
| 1748 |
-
"loss":
|
| 1749 |
"step": 4740
|
| 1750 |
},
|
| 1751 |
{
|
| 1752 |
-
"epoch":
|
| 1753 |
-
"grad_norm":
|
| 1754 |
-
"learning_rate":
|
| 1755 |
-
"loss":
|
| 1756 |
"step": 4760
|
| 1757 |
},
|
| 1758 |
{
|
| 1759 |
-
"epoch":
|
| 1760 |
-
"grad_norm":
|
| 1761 |
-
"learning_rate":
|
| 1762 |
-
"loss":
|
| 1763 |
"step": 4780
|
| 1764 |
},
|
| 1765 |
{
|
| 1766 |
-
"epoch":
|
| 1767 |
-
"grad_norm":
|
| 1768 |
-
"learning_rate":
|
| 1769 |
-
"loss":
|
| 1770 |
"step": 4800
|
| 1771 |
},
|
| 1772 |
{
|
| 1773 |
-
"epoch":
|
| 1774 |
-
"grad_norm":
|
| 1775 |
-
"learning_rate":
|
| 1776 |
-
"loss":
|
| 1777 |
"step": 4820
|
| 1778 |
},
|
| 1779 |
{
|
| 1780 |
-
"epoch":
|
| 1781 |
-
"grad_norm":
|
| 1782 |
-
"learning_rate":
|
| 1783 |
-
"loss":
|
| 1784 |
"step": 4840
|
| 1785 |
},
|
| 1786 |
{
|
| 1787 |
-
"epoch":
|
| 1788 |
-
"grad_norm":
|
| 1789 |
-
"learning_rate":
|
| 1790 |
-
"loss":
|
| 1791 |
"step": 4860
|
| 1792 |
},
|
| 1793 |
{
|
| 1794 |
-
"epoch":
|
| 1795 |
-
"grad_norm":
|
| 1796 |
-
"learning_rate":
|
| 1797 |
-
"loss":
|
| 1798 |
"step": 4880
|
| 1799 |
},
|
| 1800 |
{
|
| 1801 |
-
"epoch":
|
| 1802 |
-
"grad_norm":
|
| 1803 |
-
"learning_rate":
|
| 1804 |
-
"loss":
|
| 1805 |
"step": 4900
|
| 1806 |
},
|
| 1807 |
{
|
| 1808 |
-
"epoch":
|
| 1809 |
-
"grad_norm":
|
| 1810 |
-
"learning_rate":
|
| 1811 |
-
"loss":
|
| 1812 |
"step": 4920
|
| 1813 |
},
|
| 1814 |
{
|
| 1815 |
-
"epoch":
|
| 1816 |
-
"grad_norm":
|
| 1817 |
-
"learning_rate":
|
| 1818 |
-
"loss":
|
| 1819 |
"step": 4940
|
| 1820 |
},
|
| 1821 |
{
|
| 1822 |
-
"epoch":
|
| 1823 |
-
"grad_norm":
|
| 1824 |
-
"learning_rate":
|
| 1825 |
-
"loss":
|
| 1826 |
"step": 4960
|
| 1827 |
},
|
| 1828 |
{
|
| 1829 |
-
"epoch":
|
| 1830 |
-
"grad_norm":
|
| 1831 |
-
"learning_rate":
|
| 1832 |
-
"loss":
|
| 1833 |
"step": 4980
|
| 1834 |
},
|
| 1835 |
{
|
| 1836 |
-
"epoch":
|
| 1837 |
-
"grad_norm":
|
| 1838 |
-
"learning_rate":
|
| 1839 |
-
"loss":
|
| 1840 |
"step": 5000
|
| 1841 |
},
|
| 1842 |
{
|
| 1843 |
-
"epoch":
|
| 1844 |
-
"grad_norm":
|
| 1845 |
-
"learning_rate":
|
| 1846 |
-
"loss":
|
| 1847 |
"step": 5020
|
| 1848 |
},
|
| 1849 |
{
|
| 1850 |
-
"epoch":
|
| 1851 |
-
"grad_norm":
|
| 1852 |
-
"learning_rate":
|
| 1853 |
-
"loss":
|
| 1854 |
"step": 5040
|
| 1855 |
},
|
| 1856 |
{
|
| 1857 |
-
"epoch":
|
| 1858 |
-
"grad_norm":
|
| 1859 |
-
"learning_rate":
|
| 1860 |
-
"loss":
|
| 1861 |
"step": 5060
|
| 1862 |
},
|
| 1863 |
{
|
| 1864 |
-
"epoch":
|
| 1865 |
-
"grad_norm":
|
| 1866 |
-
"learning_rate":
|
| 1867 |
-
"loss":
|
| 1868 |
"step": 5080
|
| 1869 |
},
|
| 1870 |
{
|
| 1871 |
-
"epoch":
|
| 1872 |
-
"grad_norm":
|
| 1873 |
-
"learning_rate":
|
| 1874 |
-
"loss":
|
| 1875 |
"step": 5100
|
| 1876 |
},
|
| 1877 |
{
|
| 1878 |
-
"epoch":
|
| 1879 |
-
"grad_norm":
|
| 1880 |
-
"learning_rate":
|
| 1881 |
-
"loss":
|
| 1882 |
"step": 5120
|
| 1883 |
},
|
| 1884 |
{
|
| 1885 |
-
"epoch":
|
| 1886 |
-
"grad_norm":
|
| 1887 |
-
"learning_rate":
|
| 1888 |
-
"loss":
|
| 1889 |
"step": 5140
|
| 1890 |
},
|
| 1891 |
{
|
| 1892 |
-
"epoch":
|
| 1893 |
-
"grad_norm":
|
| 1894 |
-
"learning_rate":
|
| 1895 |
-
"loss":
|
| 1896 |
"step": 5160
|
| 1897 |
},
|
| 1898 |
{
|
| 1899 |
-
"epoch": 9.
|
| 1900 |
-
"
|
| 1901 |
-
"
|
| 1902 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1903 |
"step": 5180
|
| 1904 |
},
|
| 1905 |
{
|
| 1906 |
-
"epoch": 9.
|
| 1907 |
-
"grad_norm":
|
| 1908 |
-
"learning_rate":
|
| 1909 |
-
"loss":
|
| 1910 |
"step": 5200
|
| 1911 |
},
|
| 1912 |
{
|
| 1913 |
-
"epoch": 9.
|
| 1914 |
-
"grad_norm":
|
| 1915 |
-
"learning_rate":
|
| 1916 |
-
"loss":
|
| 1917 |
"step": 5220
|
| 1918 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1919 |
{
|
| 1920 |
"epoch": 10.0,
|
| 1921 |
-
"eval_accuracy": 0.
|
| 1922 |
-
"eval_loss":
|
| 1923 |
-
"eval_runtime":
|
| 1924 |
-
"eval_samples_per_second":
|
| 1925 |
-
"eval_steps_per_second":
|
| 1926 |
-
"step":
|
| 1927 |
},
|
| 1928 |
{
|
| 1929 |
"epoch": 10.0,
|
| 1930 |
-
"step":
|
| 1931 |
-
"total_flos": 2.
|
| 1932 |
-
"train_loss":
|
| 1933 |
-
"train_runtime":
|
| 1934 |
-
"train_samples_per_second":
|
| 1935 |
-
"train_steps_per_second": 0.
|
| 1936 |
}
|
| 1937 |
],
|
| 1938 |
"logging_steps": 20,
|
| 1939 |
-
"max_steps":
|
| 1940 |
"num_input_tokens_seen": 0,
|
| 1941 |
"num_train_epochs": 10,
|
| 1942 |
"save_steps": 500,
|
|
@@ -1952,7 +2134,7 @@
|
|
| 1952 |
"attributes": {}
|
| 1953 |
}
|
| 1954 |
},
|
| 1955 |
-
"total_flos": 2.
|
| 1956 |
"train_batch_size": 256,
|
| 1957 |
"trial_name": null,
|
| 1958 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9757901815736382,
|
| 3 |
+
"best_model_checkpoint": "/mnt/data4_HDD_14TB/yang/voxceleb-checkpoints/ecapa-tdnn/voxceleb1/pretrain/c512-aam-len3-bs256-lr5e-4/checkpoint-3450",
|
| 4 |
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 5750,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.034782608695652174,
|
| 13 |
+
"grad_norm": 6.155358791351318,
|
| 14 |
+
"learning_rate": 1.739130434782609e-05,
|
| 15 |
+
"loss": 13.2026,
|
| 16 |
"step": 20
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.06956521739130435,
|
| 20 |
+
"grad_norm": 5.816741943359375,
|
| 21 |
+
"learning_rate": 3.478260869565218e-05,
|
| 22 |
+
"loss": 13.1252,
|
| 23 |
"step": 40
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.10434782608695652,
|
| 27 |
+
"grad_norm": 5.273156642913818,
|
| 28 |
+
"learning_rate": 5.2173913043478256e-05,
|
| 29 |
+
"loss": 13.0001,
|
| 30 |
"step": 60
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.1391304347826087,
|
| 34 |
+
"grad_norm": 4.86655330657959,
|
| 35 |
+
"learning_rate": 6.956521739130436e-05,
|
| 36 |
+
"loss": 12.8639,
|
| 37 |
"step": 80
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.17391304347826086,
|
| 41 |
+
"grad_norm": 4.438321113586426,
|
| 42 |
+
"learning_rate": 8.695652173913044e-05,
|
| 43 |
+
"loss": 12.7376,
|
| 44 |
"step": 100
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.20869565217391303,
|
| 48 |
+
"grad_norm": 4.164404392242432,
|
| 49 |
+
"learning_rate": 0.00010434782608695651,
|
| 50 |
+
"loss": 12.5722,
|
| 51 |
"step": 120
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.24347826086956523,
|
| 55 |
+
"grad_norm": 3.858990430831909,
|
| 56 |
+
"learning_rate": 0.00012173913043478261,
|
| 57 |
+
"loss": 12.4229,
|
| 58 |
"step": 140
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.2782608695652174,
|
| 62 |
+
"grad_norm": 3.6574394702911377,
|
| 63 |
+
"learning_rate": 0.0001391304347826087,
|
| 64 |
+
"loss": 12.2581,
|
| 65 |
"step": 160
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.3130434782608696,
|
| 69 |
+
"grad_norm": 3.3787951469421387,
|
| 70 |
+
"learning_rate": 0.0001565217391304348,
|
| 71 |
+
"loss": 12.0753,
|
| 72 |
"step": 180
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.34782608695652173,
|
| 76 |
+
"grad_norm": 3.323820114135742,
|
| 77 |
+
"learning_rate": 0.00017391304347826088,
|
| 78 |
+
"loss": 11.9261,
|
| 79 |
"step": 200
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.3826086956521739,
|
| 83 |
+
"grad_norm": 3.247619152069092,
|
| 84 |
+
"learning_rate": 0.00019130434782608697,
|
| 85 |
+
"loss": 11.7417,
|
| 86 |
"step": 220
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.41739130434782606,
|
| 90 |
+
"grad_norm": 3.2254152297973633,
|
| 91 |
+
"learning_rate": 0.00020869565217391303,
|
| 92 |
+
"loss": 11.5771,
|
| 93 |
"step": 240
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.45217391304347826,
|
| 97 |
+
"grad_norm": 3.1803464889526367,
|
| 98 |
+
"learning_rate": 0.00022608695652173914,
|
| 99 |
+
"loss": 11.3969,
|
| 100 |
"step": 260
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.48695652173913045,
|
| 104 |
+
"grad_norm": 3.41034197807312,
|
| 105 |
+
"learning_rate": 0.00024347826086956522,
|
| 106 |
+
"loss": 11.2684,
|
| 107 |
"step": 280
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.5217391304347826,
|
| 111 |
+
"grad_norm": 3.246403217315674,
|
| 112 |
+
"learning_rate": 0.0002608695652173913,
|
| 113 |
+
"loss": 11.0744,
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.5565217391304348,
|
| 118 |
+
"grad_norm": 3.202021360397339,
|
| 119 |
+
"learning_rate": 0.0002782608695652174,
|
| 120 |
+
"loss": 10.8929,
|
| 121 |
"step": 320
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.591304347826087,
|
| 125 |
+
"grad_norm": 3.1231367588043213,
|
| 126 |
+
"learning_rate": 0.0002956521739130435,
|
| 127 |
+
"loss": 10.7468,
|
| 128 |
"step": 340
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.6260869565217392,
|
| 132 |
+
"grad_norm": 3.1820390224456787,
|
| 133 |
+
"learning_rate": 0.0003130434782608696,
|
| 134 |
+
"loss": 10.606,
|
| 135 |
"step": 360
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.6608695652173913,
|
| 139 |
+
"grad_norm": 3.2470555305480957,
|
| 140 |
+
"learning_rate": 0.0003304347826086956,
|
| 141 |
+
"loss": 10.4871,
|
| 142 |
"step": 380
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.6956521739130435,
|
| 146 |
+
"grad_norm": 3.2452709674835205,
|
| 147 |
+
"learning_rate": 0.00034782608695652176,
|
| 148 |
+
"loss": 10.2836,
|
| 149 |
"step": 400
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.7304347826086957,
|
| 153 |
+
"grad_norm": 3.203894853591919,
|
| 154 |
+
"learning_rate": 0.00036521739130434785,
|
| 155 |
+
"loss": 10.1154,
|
| 156 |
"step": 420
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.7652173913043478,
|
| 160 |
+
"grad_norm": 3.269970178604126,
|
| 161 |
+
"learning_rate": 0.00038260869565217393,
|
| 162 |
+
"loss": 9.9283,
|
| 163 |
"step": 440
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.8,
|
| 167 |
+
"grad_norm": 3.261357545852661,
|
| 168 |
+
"learning_rate": 0.0004,
|
| 169 |
+
"loss": 9.8674,
|
| 170 |
"step": 460
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.8347826086956521,
|
| 174 |
+
"grad_norm": 3.393953323364258,
|
| 175 |
+
"learning_rate": 0.00041739130434782605,
|
| 176 |
+
"loss": 9.6224,
|
| 177 |
"step": 480
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.8695652173913043,
|
| 181 |
+
"grad_norm": 3.321411609649658,
|
| 182 |
+
"learning_rate": 0.0004347826086956522,
|
| 183 |
+
"loss": 9.524,
|
| 184 |
"step": 500
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.9043478260869565,
|
| 188 |
+
"grad_norm": 3.3886823654174805,
|
| 189 |
+
"learning_rate": 0.0004521739130434783,
|
| 190 |
+
"loss": 9.384,
|
| 191 |
"step": 520
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.9391304347826087,
|
| 195 |
+
"grad_norm": 3.4735491275787354,
|
| 196 |
+
"learning_rate": 0.00046956521739130436,
|
| 197 |
+
"loss": 9.1767,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 198 |
"step": 540
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.9739130434782609,
|
| 202 |
+
"grad_norm": 3.416966676712036,
|
| 203 |
+
"learning_rate": 0.00048695652173913045,
|
| 204 |
+
"loss": 9.047,
|
| 205 |
"step": 560
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 1.0,
|
| 209 |
+
"eval_accuracy": 0.43039677202420984,
|
| 210 |
+
"eval_loss": 8.366157531738281,
|
| 211 |
+
"eval_runtime": 42.3364,
|
| 212 |
+
"eval_samples_per_second": 35.123,
|
| 213 |
+
"eval_steps_per_second": 35.123,
|
| 214 |
+
"step": 575
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.008695652173913,
|
| 218 |
+
"grad_norm": 3.446899890899658,
|
| 219 |
+
"learning_rate": 0.0004995169082125604,
|
| 220 |
+
"loss": 8.8835,
|
| 221 |
"step": 580
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 1.0434782608695652,
|
| 225 |
+
"grad_norm": 3.5842247009277344,
|
| 226 |
+
"learning_rate": 0.0004975845410628019,
|
| 227 |
+
"loss": 8.6436,
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
+
"epoch": 1.0782608695652174,
|
| 232 |
+
"grad_norm": 3.5029306411743164,
|
| 233 |
+
"learning_rate": 0.0004956521739130435,
|
| 234 |
+
"loss": 8.4775,
|
| 235 |
"step": 620
|
| 236 |
},
|
| 237 |
{
|
| 238 |
+
"epoch": 1.1130434782608696,
|
| 239 |
+
"grad_norm": 3.5451033115386963,
|
| 240 |
+
"learning_rate": 0.0004937198067632851,
|
| 241 |
+
"loss": 8.322,
|
| 242 |
"step": 640
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 1.1478260869565218,
|
| 246 |
+
"grad_norm": 3.5502634048461914,
|
| 247 |
+
"learning_rate": 0.0004917874396135266,
|
| 248 |
+
"loss": 8.1264,
|
| 249 |
"step": 660
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 1.182608695652174,
|
| 253 |
+
"grad_norm": 3.607395648956299,
|
| 254 |
+
"learning_rate": 0.0004898550724637681,
|
| 255 |
+
"loss": 7.9905,
|
| 256 |
"step": 680
|
| 257 |
},
|
| 258 |
{
|
| 259 |
+
"epoch": 1.2173913043478262,
|
| 260 |
+
"grad_norm": 3.6438565254211426,
|
| 261 |
+
"learning_rate": 0.0004879227053140097,
|
| 262 |
+
"loss": 7.8252,
|
| 263 |
"step": 700
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 1.2521739130434781,
|
| 267 |
+
"grad_norm": 3.656705141067505,
|
| 268 |
+
"learning_rate": 0.0004859903381642512,
|
| 269 |
+
"loss": 7.7737,
|
| 270 |
"step": 720
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"epoch": 1.2869565217391306,
|
| 274 |
+
"grad_norm": 3.7424328327178955,
|
| 275 |
+
"learning_rate": 0.0004840579710144928,
|
| 276 |
+
"loss": 7.5822,
|
| 277 |
"step": 740
|
| 278 |
},
|
| 279 |
{
|
| 280 |
+
"epoch": 1.3217391304347825,
|
| 281 |
+
"grad_norm": 3.673156261444092,
|
| 282 |
+
"learning_rate": 0.0004821256038647343,
|
| 283 |
+
"loss": 7.4563,
|
| 284 |
"step": 760
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 1.3565217391304347,
|
| 288 |
+
"grad_norm": 3.6774067878723145,
|
| 289 |
+
"learning_rate": 0.0004801932367149758,
|
| 290 |
+
"loss": 7.3379,
|
| 291 |
"step": 780
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 1.391304347826087,
|
| 295 |
+
"grad_norm": 3.811283826828003,
|
| 296 |
+
"learning_rate": 0.0004782608695652174,
|
| 297 |
+
"loss": 7.1559,
|
| 298 |
"step": 800
|
| 299 |
},
|
| 300 |
{
|
| 301 |
+
"epoch": 1.4260869565217391,
|
| 302 |
+
"grad_norm": 3.7899839878082275,
|
| 303 |
+
"learning_rate": 0.00047632850241545894,
|
| 304 |
+
"loss": 7.0834,
|
| 305 |
"step": 820
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 1.4608695652173913,
|
| 309 |
+
"grad_norm": 3.583247423171997,
|
| 310 |
+
"learning_rate": 0.00047439613526570047,
|
| 311 |
+
"loss": 6.9172,
|
| 312 |
"step": 840
|
| 313 |
},
|
| 314 |
{
|
| 315 |
+
"epoch": 1.4956521739130435,
|
| 316 |
+
"grad_norm": 3.8192331790924072,
|
| 317 |
+
"learning_rate": 0.00047246376811594206,
|
| 318 |
+
"loss": 6.7251,
|
| 319 |
"step": 860
|
| 320 |
},
|
| 321 |
{
|
| 322 |
+
"epoch": 1.5304347826086957,
|
| 323 |
+
"grad_norm": 3.8098299503326416,
|
| 324 |
+
"learning_rate": 0.0004705314009661836,
|
| 325 |
+
"loss": 6.7871,
|
| 326 |
"step": 880
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 1.5652173913043477,
|
| 330 |
+
"grad_norm": 3.7341325283050537,
|
| 331 |
+
"learning_rate": 0.0004685990338164252,
|
| 332 |
+
"loss": 6.6103,
|
| 333 |
"step": 900
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 1.6,
|
| 337 |
+
"grad_norm": 3.9190495014190674,
|
| 338 |
+
"learning_rate": 0.00046666666666666666,
|
| 339 |
+
"loss": 6.4507,
|
| 340 |
"step": 920
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"epoch": 1.634782608695652,
|
| 344 |
+
"grad_norm": 3.9456422328948975,
|
| 345 |
+
"learning_rate": 0.0004647342995169082,
|
| 346 |
+
"loss": 6.3619,
|
| 347 |
"step": 940
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 1.6695652173913045,
|
| 351 |
+
"grad_norm": 3.899134874343872,
|
| 352 |
+
"learning_rate": 0.0004628019323671498,
|
| 353 |
+
"loss": 6.2957,
|
| 354 |
"step": 960
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 1.7043478260869565,
|
| 358 |
+
"grad_norm": 3.878810167312622,
|
| 359 |
+
"learning_rate": 0.0004608695652173913,
|
| 360 |
+
"loss": 6.1362,
|
| 361 |
"step": 980
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 1.7391304347826086,
|
| 365 |
+
"grad_norm": 3.9270784854888916,
|
| 366 |
+
"learning_rate": 0.00045893719806763285,
|
| 367 |
+
"loss": 5.9814,
|
| 368 |
"step": 1000
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 1.7739130434782608,
|
| 372 |
+
"grad_norm": 3.8247644901275635,
|
| 373 |
+
"learning_rate": 0.00045700483091787444,
|
| 374 |
+
"loss": 5.9095,
|
| 375 |
"step": 1020
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 1.808695652173913,
|
| 379 |
+
"grad_norm": 3.8870134353637695,
|
| 380 |
+
"learning_rate": 0.000455072463768116,
|
| 381 |
+
"loss": 5.7793,
|
| 382 |
"step": 1040
|
| 383 |
},
|
| 384 |
{
|
| 385 |
+
"epoch": 1.8434782608695652,
|
| 386 |
+
"grad_norm": 3.9533441066741943,
|
| 387 |
+
"learning_rate": 0.00045314009661835745,
|
| 388 |
+
"loss": 5.7754,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 389 |
"step": 1060
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 1.8782608695652174,
|
| 393 |
+
"grad_norm": 3.9928998947143555,
|
| 394 |
+
"learning_rate": 0.00045120772946859904,
|
| 395 |
+
"loss": 5.5886,
|
| 396 |
"step": 1080
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 1.9130434782608696,
|
| 400 |
+
"grad_norm": 4.030064582824707,
|
| 401 |
+
"learning_rate": 0.0004492753623188406,
|
| 402 |
+
"loss": 5.5482,
|
| 403 |
"step": 1100
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 1.9478260869565216,
|
| 407 |
+
"grad_norm": 3.961806297302246,
|
| 408 |
+
"learning_rate": 0.0004473429951690821,
|
| 409 |
+
"loss": 5.4807,
|
| 410 |
"step": 1120
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 1.982608695652174,
|
| 414 |
+
"grad_norm": 4.003119945526123,
|
| 415 |
+
"learning_rate": 0.0004454106280193237,
|
| 416 |
+
"loss": 5.3508,
|
| 417 |
"step": 1140
|
| 418 |
},
|
| 419 |
{
|
| 420 |
+
"epoch": 2.0,
|
| 421 |
+
"eval_accuracy": 0.8190988567585743,
|
| 422 |
+
"eval_loss": 4.025164604187012,
|
| 423 |
+
"eval_runtime": 42.7144,
|
| 424 |
+
"eval_samples_per_second": 34.813,
|
| 425 |
+
"eval_steps_per_second": 34.813,
|
| 426 |
+
"step": 1150
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 2.017391304347826,
|
| 430 |
+
"grad_norm": 3.958116292953491,
|
| 431 |
+
"learning_rate": 0.00044347826086956523,
|
| 432 |
+
"loss": 5.1229,
|
| 433 |
"step": 1160
|
| 434 |
},
|
| 435 |
{
|
| 436 |
+
"epoch": 2.0521739130434784,
|
| 437 |
+
"grad_norm": 3.864279270172119,
|
| 438 |
+
"learning_rate": 0.00044154589371980677,
|
| 439 |
+
"loss": 4.8146,
|
| 440 |
"step": 1180
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"epoch": 2.0869565217391304,
|
| 444 |
+
"grad_norm": 4.045077323913574,
|
| 445 |
+
"learning_rate": 0.0004396135265700483,
|
| 446 |
+
"loss": 4.8843,
|
| 447 |
"step": 1200
|
| 448 |
},
|
| 449 |
{
|
| 450 |
+
"epoch": 2.121739130434783,
|
| 451 |
+
"grad_norm": 4.061978816986084,
|
| 452 |
+
"learning_rate": 0.00043768115942028983,
|
| 453 |
+
"loss": 4.8078,
|
| 454 |
"step": 1220
|
| 455 |
},
|
| 456 |
{
|
| 457 |
+
"epoch": 2.1565217391304348,
|
| 458 |
+
"grad_norm": 4.040159225463867,
|
| 459 |
+
"learning_rate": 0.0004357487922705314,
|
| 460 |
+
"loss": 4.6812,
|
| 461 |
"step": 1240
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 2.1913043478260867,
|
| 465 |
+
"grad_norm": 4.234623908996582,
|
| 466 |
+
"learning_rate": 0.00043381642512077296,
|
| 467 |
+
"loss": 4.6701,
|
| 468 |
"step": 1260
|
| 469 |
},
|
| 470 |
{
|
| 471 |
+
"epoch": 2.226086956521739,
|
| 472 |
+
"grad_norm": 4.030038356781006,
|
| 473 |
+
"learning_rate": 0.0004318840579710145,
|
| 474 |
+
"loss": 4.6221,
|
| 475 |
"step": 1280
|
| 476 |
},
|
| 477 |
{
|
| 478 |
+
"epoch": 2.260869565217391,
|
| 479 |
+
"grad_norm": 3.9954497814178467,
|
| 480 |
+
"learning_rate": 0.0004299516908212561,
|
| 481 |
+
"loss": 4.5647,
|
| 482 |
"step": 1300
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 2.2956521739130435,
|
| 486 |
+
"grad_norm": 4.188636779785156,
|
| 487 |
+
"learning_rate": 0.0004280193236714976,
|
| 488 |
+
"loss": 4.4502,
|
| 489 |
"step": 1320
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 2.3304347826086955,
|
| 493 |
+
"grad_norm": 4.185456275939941,
|
| 494 |
+
"learning_rate": 0.00042608695652173915,
|
| 495 |
+
"loss": 4.359,
|
| 496 |
"step": 1340
|
| 497 |
},
|
| 498 |
{
|
| 499 |
+
"epoch": 2.365217391304348,
|
| 500 |
+
"grad_norm": 4.123263359069824,
|
| 501 |
+
"learning_rate": 0.0004241545893719807,
|
| 502 |
+
"loss": 4.2863,
|
| 503 |
"step": 1360
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"epoch": 2.4,
|
| 507 |
+
"grad_norm": 4.194387435913086,
|
| 508 |
+
"learning_rate": 0.0004222222222222222,
|
| 509 |
+
"loss": 4.3354,
|
| 510 |
"step": 1380
|
| 511 |
},
|
| 512 |
{
|
| 513 |
+
"epoch": 2.4347826086956523,
|
| 514 |
+
"grad_norm": 4.065763473510742,
|
| 515 |
+
"learning_rate": 0.00042028985507246375,
|
| 516 |
+
"loss": 4.2176,
|
| 517 |
"step": 1400
|
| 518 |
},
|
| 519 |
{
|
| 520 |
+
"epoch": 2.4695652173913043,
|
| 521 |
+
"grad_norm": 4.120363712310791,
|
| 522 |
+
"learning_rate": 0.00041835748792270534,
|
| 523 |
+
"loss": 4.0597,
|
| 524 |
"step": 1420
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 2.5043478260869563,
|
| 528 |
+
"grad_norm": 4.3197174072265625,
|
| 529 |
+
"learning_rate": 0.00041642512077294687,
|
| 530 |
+
"loss": 4.028,
|
| 531 |
"step": 1440
|
| 532 |
},
|
| 533 |
{
|
| 534 |
+
"epoch": 2.5391304347826087,
|
| 535 |
+
"grad_norm": 4.2683610916137695,
|
| 536 |
+
"learning_rate": 0.0004144927536231884,
|
| 537 |
+
"loss": 3.9833,
|
| 538 |
"step": 1460
|
| 539 |
},
|
| 540 |
{
|
| 541 |
+
"epoch": 2.573913043478261,
|
| 542 |
+
"grad_norm": 4.15448522567749,
|
| 543 |
+
"learning_rate": 0.00041256038647343,
|
| 544 |
+
"loss": 4.0065,
|
| 545 |
"step": 1480
|
| 546 |
},
|
| 547 |
{
|
| 548 |
+
"epoch": 2.608695652173913,
|
| 549 |
+
"grad_norm": 4.348177433013916,
|
| 550 |
+
"learning_rate": 0.0004106280193236715,
|
| 551 |
+
"loss": 3.8134,
|
| 552 |
"step": 1500
|
| 553 |
},
|
| 554 |
{
|
| 555 |
+
"epoch": 2.643478260869565,
|
| 556 |
+
"grad_norm": 4.100021839141846,
|
| 557 |
+
"learning_rate": 0.00040869565217391306,
|
| 558 |
+
"loss": 3.8548,
|
| 559 |
"step": 1520
|
| 560 |
},
|
| 561 |
{
|
| 562 |
+
"epoch": 2.6782608695652175,
|
| 563 |
+
"grad_norm": 4.344174385070801,
|
| 564 |
+
"learning_rate": 0.0004067632850241546,
|
| 565 |
+
"loss": 3.7814,
|
| 566 |
"step": 1540
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 2.7130434782608694,
|
| 570 |
+
"grad_norm": 4.240079402923584,
|
| 571 |
+
"learning_rate": 0.00040483091787439613,
|
| 572 |
+
"loss": 3.7578,
|
| 573 |
"step": 1560
|
| 574 |
},
|
| 575 |
{
|
| 576 |
+
"epoch": 2.747826086956522,
|
| 577 |
+
"grad_norm": 4.468689918518066,
|
| 578 |
+
"learning_rate": 0.0004028985507246377,
|
| 579 |
+
"loss": 3.7331,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 580 |
"step": 1580
|
| 581 |
},
|
| 582 |
{
|
| 583 |
+
"epoch": 2.782608695652174,
|
| 584 |
+
"grad_norm": 4.28464937210083,
|
| 585 |
+
"learning_rate": 0.00040096618357487925,
|
| 586 |
+
"loss": 3.6396,
|
| 587 |
"step": 1600
|
| 588 |
},
|
| 589 |
{
|
| 590 |
+
"epoch": 2.8173913043478263,
|
| 591 |
+
"grad_norm": 4.166805744171143,
|
| 592 |
+
"learning_rate": 0.0003990338164251208,
|
| 593 |
+
"loss": 3.5799,
|
| 594 |
"step": 1620
|
| 595 |
},
|
| 596 |
{
|
| 597 |
+
"epoch": 2.8521739130434782,
|
| 598 |
+
"grad_norm": 4.237683296203613,
|
| 599 |
+
"learning_rate": 0.0003971014492753624,
|
| 600 |
+
"loss": 3.4734,
|
| 601 |
"step": 1640
|
| 602 |
},
|
| 603 |
{
|
| 604 |
+
"epoch": 2.8869565217391306,
|
| 605 |
+
"grad_norm": 4.153097152709961,
|
| 606 |
+
"learning_rate": 0.00039516908212560385,
|
| 607 |
+
"loss": 3.5183,
|
| 608 |
"step": 1660
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 2.9217391304347826,
|
| 612 |
+
"grad_norm": 4.2313947677612305,
|
| 613 |
+
"learning_rate": 0.0003932367149758454,
|
| 614 |
+
"loss": 3.3963,
|
| 615 |
"step": 1680
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 2.9565217391304346,
|
| 619 |
+
"grad_norm": 3.992475748062134,
|
| 620 |
+
"learning_rate": 0.000391304347826087,
|
| 621 |
+
"loss": 3.3081,
|
| 622 |
"step": 1700
|
| 623 |
},
|
| 624 |
{
|
| 625 |
+
"epoch": 2.991304347826087,
|
| 626 |
+
"grad_norm": 4.4731059074401855,
|
| 627 |
+
"learning_rate": 0.0003893719806763285,
|
| 628 |
+
"loss": 3.3124,
|
| 629 |
"step": 1720
|
| 630 |
},
|
| 631 |
{
|
| 632 |
+
"epoch": 3.0,
|
| 633 |
+
"eval_accuracy": 0.9260255548083389,
|
| 634 |
+
"eval_loss": 2.1082653999328613,
|
| 635 |
+
"eval_runtime": 22.1676,
|
| 636 |
+
"eval_samples_per_second": 67.08,
|
| 637 |
+
"eval_steps_per_second": 67.08,
|
| 638 |
+
"step": 1725
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 3.026086956521739,
|
| 642 |
+
"grad_norm": 4.272000312805176,
|
| 643 |
+
"learning_rate": 0.00038743961352657004,
|
| 644 |
+
"loss": 3.1247,
|
| 645 |
"step": 1740
|
| 646 |
},
|
| 647 |
{
|
| 648 |
+
"epoch": 3.0608695652173914,
|
| 649 |
+
"grad_norm": 4.102330207824707,
|
| 650 |
+
"learning_rate": 0.00038550724637681163,
|
| 651 |
+
"loss": 3.1064,
|
| 652 |
"step": 1760
|
| 653 |
},
|
| 654 |
{
|
| 655 |
+
"epoch": 3.0956521739130434,
|
| 656 |
+
"grad_norm": 4.381846904754639,
|
| 657 |
+
"learning_rate": 0.00038357487922705317,
|
| 658 |
+
"loss": 2.9371,
|
| 659 |
"step": 1780
|
| 660 |
},
|
| 661 |
{
|
| 662 |
+
"epoch": 3.130434782608696,
|
| 663 |
+
"grad_norm": 4.1588921546936035,
|
| 664 |
+
"learning_rate": 0.00038164251207729465,
|
| 665 |
+
"loss": 2.9355,
|
| 666 |
"step": 1800
|
| 667 |
},
|
| 668 |
{
|
| 669 |
+
"epoch": 3.1652173913043478,
|
| 670 |
+
"grad_norm": 4.279609203338623,
|
| 671 |
+
"learning_rate": 0.00037971014492753623,
|
| 672 |
+
"loss": 2.8545,
|
| 673 |
"step": 1820
|
| 674 |
},
|
| 675 |
{
|
| 676 |
+
"epoch": 3.2,
|
| 677 |
+
"grad_norm": 4.240756988525391,
|
| 678 |
+
"learning_rate": 0.00037777777777777777,
|
| 679 |
+
"loss": 2.8096,
|
| 680 |
"step": 1840
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"epoch": 3.234782608695652,
|
| 684 |
+
"grad_norm": 4.11091947555542,
|
| 685 |
+
"learning_rate": 0.00037584541062801936,
|
| 686 |
+
"loss": 2.8138,
|
| 687 |
"step": 1860
|
| 688 |
},
|
| 689 |
{
|
| 690 |
+
"epoch": 3.269565217391304,
|
| 691 |
+
"grad_norm": 4.078794479370117,
|
| 692 |
+
"learning_rate": 0.0003739130434782609,
|
| 693 |
+
"loss": 2.7417,
|
| 694 |
"step": 1880
|
| 695 |
},
|
| 696 |
{
|
| 697 |
+
"epoch": 3.3043478260869565,
|
| 698 |
+
"grad_norm": 4.368116855621338,
|
| 699 |
+
"learning_rate": 0.0003719806763285024,
|
| 700 |
+
"loss": 2.7937,
|
| 701 |
"step": 1900
|
| 702 |
},
|
| 703 |
{
|
| 704 |
+
"epoch": 3.3391304347826085,
|
| 705 |
+
"grad_norm": 4.044319152832031,
|
| 706 |
+
"learning_rate": 0.000370048309178744,
|
| 707 |
+
"loss": 2.7361,
|
| 708 |
"step": 1920
|
| 709 |
},
|
| 710 |
{
|
| 711 |
+
"epoch": 3.373913043478261,
|
| 712 |
+
"grad_norm": 4.314040184020996,
|
| 713 |
+
"learning_rate": 0.0003681159420289855,
|
| 714 |
+
"loss": 2.7054,
|
| 715 |
"step": 1940
|
| 716 |
},
|
| 717 |
{
|
| 718 |
+
"epoch": 3.408695652173913,
|
| 719 |
+
"grad_norm": 4.185855388641357,
|
| 720 |
+
"learning_rate": 0.000366183574879227,
|
| 721 |
+
"loss": 2.6682,
|
| 722 |
"step": 1960
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 3.4434782608695653,
|
| 726 |
+
"grad_norm": 4.433622360229492,
|
| 727 |
+
"learning_rate": 0.0003642512077294686,
|
| 728 |
+
"loss": 2.6644,
|
| 729 |
"step": 1980
|
| 730 |
},
|
| 731 |
{
|
| 732 |
+
"epoch": 3.4782608695652173,
|
| 733 |
+
"grad_norm": 4.048947811126709,
|
| 734 |
+
"learning_rate": 0.00036231884057971015,
|
| 735 |
+
"loss": 2.618,
|
| 736 |
"step": 2000
|
| 737 |
},
|
| 738 |
{
|
| 739 |
+
"epoch": 3.5130434782608697,
|
| 740 |
+
"grad_norm": 4.145406246185303,
|
| 741 |
+
"learning_rate": 0.0003603864734299517,
|
| 742 |
+
"loss": 2.5982,
|
| 743 |
"step": 2020
|
| 744 |
},
|
| 745 |
{
|
| 746 |
+
"epoch": 3.5478260869565217,
|
| 747 |
+
"grad_norm": 4.2812910079956055,
|
| 748 |
+
"learning_rate": 0.00035845410628019327,
|
| 749 |
+
"loss": 2.6138,
|
| 750 |
"step": 2040
|
| 751 |
},
|
| 752 |
{
|
| 753 |
+
"epoch": 3.5826086956521737,
|
| 754 |
+
"grad_norm": 4.400162220001221,
|
| 755 |
+
"learning_rate": 0.0003565217391304348,
|
| 756 |
+
"loss": 2.5039,
|
| 757 |
"step": 2060
|
| 758 |
},
|
| 759 |
{
|
| 760 |
+
"epoch": 3.617391304347826,
|
| 761 |
+
"grad_norm": 4.217800617218018,
|
| 762 |
+
"learning_rate": 0.0003545893719806763,
|
| 763 |
+
"loss": 2.5249,
|
| 764 |
"step": 2080
|
| 765 |
},
|
| 766 |
{
|
| 767 |
+
"epoch": 3.6521739130434785,
|
| 768 |
+
"grad_norm": 4.076215744018555,
|
| 769 |
+
"learning_rate": 0.0003526570048309179,
|
| 770 |
+
"loss": 2.4547,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 771 |
"step": 2100
|
| 772 |
},
|
| 773 |
{
|
| 774 |
+
"epoch": 3.6869565217391305,
|
| 775 |
+
"grad_norm": 4.139514446258545,
|
| 776 |
+
"learning_rate": 0.0003507246376811594,
|
| 777 |
+
"loss": 2.4315,
|
| 778 |
"step": 2120
|
| 779 |
},
|
| 780 |
{
|
| 781 |
+
"epoch": 3.7217391304347824,
|
| 782 |
+
"grad_norm": 4.118022918701172,
|
| 783 |
+
"learning_rate": 0.00034879227053140094,
|
| 784 |
+
"loss": 2.3836,
|
| 785 |
"step": 2140
|
| 786 |
},
|
| 787 |
{
|
| 788 |
+
"epoch": 3.756521739130435,
|
| 789 |
+
"grad_norm": 4.137601852416992,
|
| 790 |
+
"learning_rate": 0.00034685990338164253,
|
| 791 |
+
"loss": 2.3284,
|
| 792 |
"step": 2160
|
| 793 |
},
|
| 794 |
{
|
| 795 |
+
"epoch": 3.791304347826087,
|
| 796 |
+
"grad_norm": 4.023979663848877,
|
| 797 |
+
"learning_rate": 0.00034492753623188406,
|
| 798 |
+
"loss": 2.3095,
|
| 799 |
"step": 2180
|
| 800 |
},
|
| 801 |
{
|
| 802 |
+
"epoch": 3.8260869565217392,
|
| 803 |
+
"grad_norm": 4.042725086212158,
|
| 804 |
+
"learning_rate": 0.00034299516908212565,
|
| 805 |
+
"loss": 2.305,
|
| 806 |
"step": 2200
|
| 807 |
},
|
| 808 |
{
|
| 809 |
+
"epoch": 3.860869565217391,
|
| 810 |
+
"grad_norm": 4.265875339508057,
|
| 811 |
+
"learning_rate": 0.0003410628019323672,
|
| 812 |
+
"loss": 2.3237,
|
| 813 |
"step": 2220
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"epoch": 3.8956521739130436,
|
| 817 |
+
"grad_norm": 4.205041408538818,
|
| 818 |
+
"learning_rate": 0.00033913043478260867,
|
| 819 |
+
"loss": 2.335,
|
| 820 |
"step": 2240
|
| 821 |
},
|
| 822 |
{
|
| 823 |
+
"epoch": 3.9304347826086956,
|
| 824 |
+
"grad_norm": 4.1344709396362305,
|
| 825 |
+
"learning_rate": 0.00033719806763285025,
|
| 826 |
+
"loss": 2.2341,
|
| 827 |
"step": 2260
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 3.965217391304348,
|
| 831 |
+
"grad_norm": 4.247790813446045,
|
| 832 |
+
"learning_rate": 0.0003352657004830918,
|
| 833 |
+
"loss": 2.251,
|
| 834 |
"step": 2280
|
| 835 |
},
|
| 836 |
{
|
| 837 |
+
"epoch": 4.0,
|
| 838 |
+
"grad_norm": 4.859626770019531,
|
| 839 |
+
"learning_rate": 0.0003333333333333333,
|
| 840 |
+
"loss": 2.3212,
|
| 841 |
+
"step": 2300
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 4.0,
|
| 845 |
+
"eval_accuracy": 0.9435104236718225,
|
| 846 |
+
"eval_loss": 1.2223739624023438,
|
| 847 |
+
"eval_runtime": 14.8513,
|
| 848 |
+
"eval_samples_per_second": 100.126,
|
| 849 |
+
"eval_steps_per_second": 100.126,
|
| 850 |
"step": 2300
|
| 851 |
},
|
| 852 |
{
|
| 853 |
+
"epoch": 4.034782608695652,
|
| 854 |
+
"grad_norm": 4.098020553588867,
|
| 855 |
+
"learning_rate": 0.0003314009661835749,
|
| 856 |
+
"loss": 1.9133,
|
| 857 |
"step": 2320
|
| 858 |
},
|
| 859 |
{
|
| 860 |
+
"epoch": 4.069565217391304,
|
| 861 |
+
"grad_norm": 4.198029041290283,
|
| 862 |
+
"learning_rate": 0.00032946859903381644,
|
| 863 |
+
"loss": 1.9814,
|
| 864 |
"step": 2340
|
| 865 |
},
|
| 866 |
{
|
| 867 |
+
"epoch": 4.104347826086957,
|
| 868 |
+
"grad_norm": 3.960844039916992,
|
| 869 |
+
"learning_rate": 0.000327536231884058,
|
| 870 |
+
"loss": 1.9505,
|
| 871 |
"step": 2360
|
| 872 |
},
|
| 873 |
{
|
| 874 |
+
"epoch": 4.139130434782609,
|
| 875 |
+
"grad_norm": 4.0190300941467285,
|
| 876 |
+
"learning_rate": 0.0003256038647342995,
|
| 877 |
+
"loss": 1.8815,
|
| 878 |
"step": 2380
|
| 879 |
},
|
| 880 |
{
|
| 881 |
+
"epoch": 4.173913043478261,
|
| 882 |
+
"grad_norm": 4.040708541870117,
|
| 883 |
+
"learning_rate": 0.00032367149758454105,
|
| 884 |
+
"loss": 1.8365,
|
| 885 |
"step": 2400
|
| 886 |
},
|
| 887 |
{
|
| 888 |
+
"epoch": 4.208695652173913,
|
| 889 |
+
"grad_norm": 4.077364444732666,
|
| 890 |
+
"learning_rate": 0.0003217391304347826,
|
| 891 |
+
"loss": 1.84,
|
| 892 |
"step": 2420
|
| 893 |
},
|
| 894 |
{
|
| 895 |
+
"epoch": 4.243478260869566,
|
| 896 |
+
"grad_norm": 4.267309188842773,
|
| 897 |
+
"learning_rate": 0.0003199033816425121,
|
| 898 |
+
"loss": 1.8864,
|
| 899 |
"step": 2440
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 4.278260869565218,
|
| 903 |
+
"grad_norm": 3.978663921356201,
|
| 904 |
+
"learning_rate": 0.00031797101449275363,
|
| 905 |
+
"loss": 1.9015,
|
| 906 |
"step": 2460
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 4.3130434782608695,
|
| 910 |
+
"grad_norm": 4.089256763458252,
|
| 911 |
+
"learning_rate": 0.0003160386473429952,
|
| 912 |
+
"loss": 1.8388,
|
| 913 |
"step": 2480
|
| 914 |
},
|
| 915 |
{
|
| 916 |
+
"epoch": 4.3478260869565215,
|
| 917 |
+
"grad_norm": 3.9317057132720947,
|
| 918 |
+
"learning_rate": 0.0003141062801932367,
|
| 919 |
+
"loss": 1.7845,
|
| 920 |
"step": 2500
|
| 921 |
},
|
| 922 |
{
|
| 923 |
+
"epoch": 4.3826086956521735,
|
| 924 |
+
"grad_norm": 3.9738080501556396,
|
| 925 |
+
"learning_rate": 0.00031217391304347823,
|
| 926 |
+
"loss": 1.7725,
|
| 927 |
"step": 2520
|
| 928 |
},
|
| 929 |
{
|
| 930 |
+
"epoch": 4.417391304347826,
|
| 931 |
+
"grad_norm": 4.232215881347656,
|
| 932 |
+
"learning_rate": 0.0003102415458937198,
|
| 933 |
+
"loss": 1.852,
|
| 934 |
"step": 2540
|
| 935 |
},
|
| 936 |
{
|
| 937 |
+
"epoch": 4.452173913043478,
|
| 938 |
+
"grad_norm": 4.050131797790527,
|
| 939 |
+
"learning_rate": 0.00030830917874396136,
|
| 940 |
+
"loss": 1.8234,
|
| 941 |
"step": 2560
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"epoch": 4.48695652173913,
|
| 945 |
+
"grad_norm": 4.217935085296631,
|
| 946 |
+
"learning_rate": 0.0003063768115942029,
|
| 947 |
+
"loss": 1.8148,
|
| 948 |
"step": 2580
|
| 949 |
},
|
| 950 |
{
|
| 951 |
+
"epoch": 4.521739130434782,
|
| 952 |
+
"grad_norm": 3.9807074069976807,
|
| 953 |
+
"learning_rate": 0.0003044444444444445,
|
| 954 |
+
"loss": 1.7134,
|
| 955 |
"step": 2600
|
| 956 |
},
|
| 957 |
{
|
| 958 |
+
"epoch": 4.556521739130435,
|
| 959 |
+
"grad_norm": 4.05940580368042,
|
| 960 |
+
"learning_rate": 0.000302512077294686,
|
| 961 |
+
"loss": 1.6752,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 962 |
"step": 2620
|
| 963 |
},
|
| 964 |
{
|
| 965 |
+
"epoch": 4.591304347826087,
|
| 966 |
+
"grad_norm": 4.454566955566406,
|
| 967 |
+
"learning_rate": 0.00030057971014492755,
|
| 968 |
+
"loss": 1.8413,
|
| 969 |
"step": 2640
|
| 970 |
},
|
| 971 |
{
|
| 972 |
+
"epoch": 4.626086956521739,
|
| 973 |
+
"grad_norm": 4.144088268280029,
|
| 974 |
+
"learning_rate": 0.0002986473429951691,
|
| 975 |
+
"loss": 1.7948,
|
| 976 |
"step": 2660
|
| 977 |
},
|
| 978 |
{
|
| 979 |
+
"epoch": 4.660869565217391,
|
| 980 |
+
"grad_norm": 3.940176010131836,
|
| 981 |
+
"learning_rate": 0.0002967149758454106,
|
| 982 |
+
"loss": 1.7468,
|
| 983 |
"step": 2680
|
| 984 |
},
|
| 985 |
{
|
| 986 |
+
"epoch": 4.695652173913043,
|
| 987 |
+
"grad_norm": 4.198675632476807,
|
| 988 |
+
"learning_rate": 0.0002948792270531401,
|
| 989 |
+
"loss": 1.709,
|
| 990 |
"step": 2700
|
| 991 |
},
|
| 992 |
{
|
| 993 |
+
"epoch": 4.730434782608696,
|
| 994 |
+
"grad_norm": 3.976001501083374,
|
| 995 |
+
"learning_rate": 0.00029294685990338167,
|
| 996 |
+
"loss": 1.6506,
|
| 997 |
"step": 2720
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
+
"epoch": 4.765217391304348,
|
| 1001 |
+
"grad_norm": 4.033059120178223,
|
| 1002 |
+
"learning_rate": 0.0002910144927536232,
|
| 1003 |
+
"loss": 1.7042,
|
| 1004 |
"step": 2740
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
+
"epoch": 4.8,
|
| 1008 |
+
"grad_norm": 4.062041759490967,
|
| 1009 |
+
"learning_rate": 0.0002890821256038648,
|
| 1010 |
+
"loss": 1.6795,
|
| 1011 |
"step": 2760
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
+
"epoch": 4.834782608695652,
|
| 1015 |
+
"grad_norm": 3.988589286804199,
|
| 1016 |
+
"learning_rate": 0.00028714975845410627,
|
| 1017 |
+
"loss": 1.7029,
|
| 1018 |
"step": 2780
|
| 1019 |
},
|
| 1020 |
{
|
| 1021 |
+
"epoch": 4.869565217391305,
|
| 1022 |
+
"grad_norm": 4.16325044631958,
|
| 1023 |
+
"learning_rate": 0.0002852173913043478,
|
| 1024 |
+
"loss": 1.6641,
|
| 1025 |
"step": 2800
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
+
"epoch": 4.904347826086957,
|
| 1029 |
+
"grad_norm": 4.323537349700928,
|
| 1030 |
+
"learning_rate": 0.0002832850241545894,
|
| 1031 |
+
"loss": 1.6953,
|
| 1032 |
"step": 2820
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
+
"epoch": 4.939130434782609,
|
| 1036 |
+
"grad_norm": 3.8293144702911377,
|
| 1037 |
+
"learning_rate": 0.0002813526570048309,
|
| 1038 |
+
"loss": 1.5863,
|
| 1039 |
"step": 2840
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
+
"epoch": 4.973913043478261,
|
| 1043 |
+
"grad_norm": 3.8955535888671875,
|
| 1044 |
+
"learning_rate": 0.00027942028985507246,
|
| 1045 |
+
"loss": 1.6276,
|
| 1046 |
"step": 2860
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
+
"epoch": 5.0,
|
| 1050 |
+
"eval_accuracy": 0.9677202420981843,
|
| 1051 |
+
"eval_loss": 0.8229038715362549,
|
| 1052 |
+
"eval_runtime": 88.6744,
|
| 1053 |
+
"eval_samples_per_second": 16.769,
|
| 1054 |
+
"eval_steps_per_second": 16.769,
|
| 1055 |
+
"step": 2875
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 5.008695652173913,
|
| 1059 |
+
"grad_norm": 3.8480091094970703,
|
| 1060 |
+
"learning_rate": 0.00027748792270531405,
|
| 1061 |
+
"loss": 1.5701,
|
| 1062 |
"step": 2880
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
+
"epoch": 5.043478260869565,
|
| 1066 |
+
"grad_norm": 3.679872512817383,
|
| 1067 |
+
"learning_rate": 0.0002755555555555556,
|
| 1068 |
+
"loss": 1.3786,
|
| 1069 |
"step": 2900
|
| 1070 |
},
|
| 1071 |
{
|
| 1072 |
+
"epoch": 5.078260869565217,
|
| 1073 |
+
"grad_norm": 4.13381290435791,
|
| 1074 |
+
"learning_rate": 0.00027362318840579706,
|
| 1075 |
+
"loss": 1.3563,
|
| 1076 |
"step": 2920
|
| 1077 |
},
|
| 1078 |
{
|
| 1079 |
+
"epoch": 5.113043478260869,
|
| 1080 |
+
"grad_norm": 3.7467329502105713,
|
| 1081 |
+
"learning_rate": 0.00027169082125603865,
|
| 1082 |
+
"loss": 1.3588,
|
| 1083 |
"step": 2940
|
| 1084 |
},
|
| 1085 |
{
|
| 1086 |
+
"epoch": 5.147826086956521,
|
| 1087 |
+
"grad_norm": 3.5837419033050537,
|
| 1088 |
+
"learning_rate": 0.0002698550724637681,
|
| 1089 |
+
"loss": 1.3782,
|
| 1090 |
"step": 2960
|
| 1091 |
},
|
| 1092 |
{
|
| 1093 |
+
"epoch": 5.182608695652174,
|
| 1094 |
+
"grad_norm": 4.077097415924072,
|
| 1095 |
+
"learning_rate": 0.00026792270531400964,
|
| 1096 |
+
"loss": 1.3969,
|
| 1097 |
"step": 2980
|
| 1098 |
},
|
| 1099 |
{
|
| 1100 |
+
"epoch": 5.217391304347826,
|
| 1101 |
+
"grad_norm": 3.5995211601257324,
|
| 1102 |
+
"learning_rate": 0.00026599033816425123,
|
| 1103 |
+
"loss": 1.3346,
|
| 1104 |
"step": 3000
|
| 1105 |
},
|
| 1106 |
{
|
| 1107 |
+
"epoch": 5.252173913043478,
|
| 1108 |
+
"grad_norm": 3.714010000228882,
|
| 1109 |
+
"learning_rate": 0.00026405797101449277,
|
| 1110 |
+
"loss": 1.3772,
|
| 1111 |
"step": 3020
|
| 1112 |
},
|
| 1113 |
{
|
| 1114 |
+
"epoch": 5.28695652173913,
|
| 1115 |
+
"grad_norm": 3.807094097137451,
|
| 1116 |
+
"learning_rate": 0.00026231884057971016,
|
| 1117 |
+
"loss": 1.3452,
|
| 1118 |
"step": 3040
|
| 1119 |
},
|
| 1120 |
{
|
| 1121 |
+
"epoch": 5.321739130434783,
|
| 1122 |
+
"grad_norm": 4.012477397918701,
|
| 1123 |
+
"learning_rate": 0.0002603864734299517,
|
| 1124 |
+
"loss": 1.3161,
|
| 1125 |
"step": 3060
|
| 1126 |
},
|
| 1127 |
{
|
| 1128 |
+
"epoch": 5.356521739130435,
|
| 1129 |
+
"grad_norm": 3.850520372390747,
|
| 1130 |
+
"learning_rate": 0.0002584541062801932,
|
| 1131 |
+
"loss": 1.3146,
|
| 1132 |
"step": 3080
|
| 1133 |
},
|
| 1134 |
{
|
| 1135 |
+
"epoch": 5.391304347826087,
|
| 1136 |
+
"grad_norm": NaN,
|
| 1137 |
+
"learning_rate": 0.00025661835748792274,
|
| 1138 |
+
"loss": 1.3057,
|
| 1139 |
"step": 3100
|
| 1140 |
},
|
| 1141 |
{
|
| 1142 |
+
"epoch": 5.426086956521739,
|
| 1143 |
+
"grad_norm": 3.697744607925415,
|
| 1144 |
+
"learning_rate": 0.0002546859903381643,
|
| 1145 |
+
"loss": 1.2619,
|
| 1146 |
"step": 3120
|
| 1147 |
},
|
| 1148 |
{
|
| 1149 |
+
"epoch": 5.460869565217392,
|
| 1150 |
+
"grad_norm": 4.125018119812012,
|
| 1151 |
+
"learning_rate": 0.00025275362318840576,
|
| 1152 |
+
"loss": 1.3436,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1153 |
"step": 3140
|
| 1154 |
},
|
| 1155 |
{
|
| 1156 |
+
"epoch": 5.495652173913044,
|
| 1157 |
+
"grad_norm": 4.1491899490356445,
|
| 1158 |
+
"learning_rate": 0.00025082125603864735,
|
| 1159 |
+
"loss": 1.3289,
|
| 1160 |
"step": 3160
|
| 1161 |
},
|
| 1162 |
{
|
| 1163 |
+
"epoch": 5.530434782608696,
|
| 1164 |
+
"grad_norm": 3.9294846057891846,
|
| 1165 |
+
"learning_rate": 0.0002488888888888889,
|
| 1166 |
+
"loss": 1.218,
|
| 1167 |
"step": 3180
|
| 1168 |
},
|
| 1169 |
{
|
| 1170 |
+
"epoch": 5.565217391304348,
|
| 1171 |
+
"grad_norm": 3.9030706882476807,
|
| 1172 |
+
"learning_rate": 0.00024695652173913047,
|
| 1173 |
+
"loss": 1.3219,
|
| 1174 |
"step": 3200
|
| 1175 |
},
|
| 1176 |
{
|
| 1177 |
+
"epoch": 5.6,
|
| 1178 |
+
"grad_norm": 4.124849319458008,
|
| 1179 |
+
"learning_rate": 0.000245024154589372,
|
| 1180 |
+
"loss": 1.2694,
|
| 1181 |
"step": 3220
|
| 1182 |
},
|
| 1183 |
{
|
| 1184 |
+
"epoch": 5.6347826086956525,
|
| 1185 |
+
"grad_norm": 4.1668500900268555,
|
| 1186 |
+
"learning_rate": 0.0002432850241545894,
|
| 1187 |
+
"loss": 1.2379,
|
| 1188 |
"step": 3240
|
| 1189 |
},
|
| 1190 |
{
|
| 1191 |
+
"epoch": 5.6695652173913045,
|
| 1192 |
+
"grad_norm": 4.098198890686035,
|
| 1193 |
+
"learning_rate": 0.00024135265700483093,
|
| 1194 |
+
"loss": 1.2892,
|
| 1195 |
"step": 3260
|
| 1196 |
},
|
| 1197 |
{
|
| 1198 |
+
"epoch": 5.7043478260869565,
|
| 1199 |
+
"grad_norm": 3.690241813659668,
|
| 1200 |
+
"learning_rate": 0.00023942028985507246,
|
| 1201 |
+
"loss": 1.2742,
|
| 1202 |
"step": 3280
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
+
"epoch": 5.739130434782608,
|
| 1206 |
+
"grad_norm": 3.978963613510132,
|
| 1207 |
+
"learning_rate": 0.00023748792270531402,
|
| 1208 |
+
"loss": 1.1755,
|
| 1209 |
"step": 3300
|
| 1210 |
},
|
| 1211 |
{
|
| 1212 |
+
"epoch": 5.773913043478261,
|
| 1213 |
+
"grad_norm": 3.7397215366363525,
|
| 1214 |
+
"learning_rate": 0.00023574879227053139,
|
| 1215 |
+
"loss": 1.2256,
|
| 1216 |
"step": 3320
|
| 1217 |
},
|
| 1218 |
{
|
| 1219 |
+
"epoch": 5.808695652173913,
|
| 1220 |
+
"grad_norm": 3.9201064109802246,
|
| 1221 |
+
"learning_rate": 0.00023391304347826088,
|
| 1222 |
+
"loss": 1.238,
|
| 1223 |
"step": 3340
|
| 1224 |
},
|
| 1225 |
{
|
| 1226 |
+
"epoch": 5.843478260869565,
|
| 1227 |
+
"grad_norm": 3.725389242172241,
|
| 1228 |
+
"learning_rate": 0.0002319806763285024,
|
| 1229 |
+
"loss": 1.1706,
|
| 1230 |
"step": 3360
|
| 1231 |
},
|
| 1232 |
{
|
| 1233 |
+
"epoch": 5.878260869565217,
|
| 1234 |
+
"grad_norm": 3.5844123363494873,
|
| 1235 |
+
"learning_rate": 0.00023004830917874397,
|
| 1236 |
+
"loss": 1.1644,
|
| 1237 |
"step": 3380
|
| 1238 |
},
|
| 1239 |
{
|
| 1240 |
+
"epoch": 5.913043478260869,
|
| 1241 |
+
"grad_norm": 3.79936146736145,
|
| 1242 |
+
"learning_rate": 0.00022821256038647343,
|
| 1243 |
+
"loss": 1.2256,
|
| 1244 |
"step": 3400
|
| 1245 |
},
|
| 1246 |
{
|
| 1247 |
+
"epoch": 5.947826086956522,
|
| 1248 |
+
"grad_norm": 3.5947725772857666,
|
| 1249 |
+
"learning_rate": 0.00022628019323671497,
|
| 1250 |
+
"loss": 1.2488,
|
| 1251 |
"step": 3420
|
| 1252 |
},
|
| 1253 |
{
|
| 1254 |
+
"epoch": 5.982608695652174,
|
| 1255 |
+
"grad_norm": NaN,
|
| 1256 |
+
"learning_rate": 0.00022444444444444446,
|
| 1257 |
+
"loss": 1.1418,
|
| 1258 |
"step": 3440
|
| 1259 |
},
|
| 1260 |
{
|
| 1261 |
+
"epoch": 6.0,
|
| 1262 |
+
"eval_accuracy": 0.9757901815736382,
|
| 1263 |
+
"eval_loss": 0.5840117335319519,
|
| 1264 |
+
"eval_runtime": 97.2696,
|
| 1265 |
+
"eval_samples_per_second": 15.287,
|
| 1266 |
+
"eval_steps_per_second": 15.287,
|
| 1267 |
+
"step": 3450
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 6.017391304347826,
|
| 1271 |
+
"grad_norm": 3.5959298610687256,
|
| 1272 |
+
"learning_rate": 0.00022260869565217392,
|
| 1273 |
+
"loss": 1.1254,
|
| 1274 |
"step": 3460
|
| 1275 |
},
|
| 1276 |
{
|
| 1277 |
+
"epoch": 6.052173913043478,
|
| 1278 |
+
"grad_norm": 3.9623775482177734,
|
| 1279 |
+
"learning_rate": 0.00022067632850241545,
|
| 1280 |
+
"loss": 1.0343,
|
| 1281 |
"step": 3480
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
+
"epoch": 6.086956521739131,
|
| 1285 |
+
"grad_norm": 3.735102415084839,
|
| 1286 |
+
"learning_rate": 0.00021874396135265702,
|
| 1287 |
+
"loss": 1.0348,
|
| 1288 |
"step": 3500
|
| 1289 |
},
|
| 1290 |
{
|
| 1291 |
+
"epoch": 6.121739130434783,
|
| 1292 |
+
"grad_norm": 3.4255013465881348,
|
| 1293 |
+
"learning_rate": 0.00021681159420289855,
|
| 1294 |
+
"loss": 0.9796,
|
| 1295 |
"step": 3520
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
+
"epoch": 6.156521739130435,
|
| 1299 |
+
"grad_norm": 3.981841564178467,
|
| 1300 |
+
"learning_rate": 0.00021497584541062804,
|
| 1301 |
+
"loss": 0.9865,
|
| 1302 |
"step": 3540
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
+
"epoch": 6.191304347826087,
|
| 1306 |
+
"grad_norm": 3.9057116508483887,
|
| 1307 |
+
"learning_rate": 0.00021314009661835748,
|
| 1308 |
+
"loss": 1.0054,
|
| 1309 |
"step": 3560
|
| 1310 |
},
|
| 1311 |
{
|
| 1312 |
+
"epoch": 6.226086956521739,
|
| 1313 |
+
"grad_norm": 3.626560688018799,
|
| 1314 |
+
"learning_rate": 0.00021120772946859904,
|
| 1315 |
+
"loss": 1.0012,
|
| 1316 |
"step": 3580
|
| 1317 |
},
|
| 1318 |
{
|
| 1319 |
+
"epoch": 6.260869565217392,
|
| 1320 |
+
"grad_norm": 3.687683582305908,
|
| 1321 |
+
"learning_rate": 0.0002093719806763285,
|
| 1322 |
+
"loss": 1.0129,
|
| 1323 |
"step": 3600
|
| 1324 |
},
|
| 1325 |
{
|
| 1326 |
+
"epoch": 6.2956521739130435,
|
| 1327 |
+
"grad_norm": 3.8632826805114746,
|
| 1328 |
+
"learning_rate": 0.00020763285024154592,
|
| 1329 |
+
"loss": 0.9333,
|
| 1330 |
"step": 3620
|
| 1331 |
},
|
| 1332 |
{
|
| 1333 |
+
"epoch": 6.3304347826086955,
|
| 1334 |
+
"grad_norm": 4.089422702789307,
|
| 1335 |
+
"learning_rate": 0.0002058937198067633,
|
| 1336 |
+
"loss": 1.0259,
|
| 1337 |
"step": 3640
|
| 1338 |
},
|
| 1339 |
{
|
| 1340 |
+
"epoch": 6.3652173913043475,
|
| 1341 |
+
"grad_norm": 4.261268615722656,
|
| 1342 |
+
"learning_rate": 0.00020415458937198067,
|
| 1343 |
+
"loss": 1.0184,
|
| 1344 |
"step": 3660
|
| 1345 |
},
|
| 1346 |
{
|
| 1347 |
+
"epoch": 6.4,
|
| 1348 |
+
"grad_norm": 2.3901586532592773,
|
| 1349 |
+
"learning_rate": 0.0002026086956521739,
|
| 1350 |
+
"loss": 1.0293,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1351 |
"step": 3680
|
| 1352 |
},
|
| 1353 |
{
|
| 1354 |
+
"epoch": 6.434782608695652,
|
| 1355 |
+
"grad_norm": 2.233633518218994,
|
| 1356 |
+
"learning_rate": 0.00020067632850241546,
|
| 1357 |
+
"loss": 1.0026,
|
| 1358 |
"step": 3700
|
| 1359 |
},
|
| 1360 |
{
|
| 1361 |
+
"epoch": 6.469565217391304,
|
| 1362 |
+
"grad_norm": 2.049773693084717,
|
| 1363 |
+
"learning_rate": 0.00019893719806763285,
|
| 1364 |
+
"loss": 1.0426,
|
| 1365 |
"step": 3720
|
| 1366 |
},
|
| 1367 |
{
|
| 1368 |
+
"epoch": 6.504347826086956,
|
| 1369 |
+
"grad_norm": 2.21939754486084,
|
| 1370 |
+
"learning_rate": 0.0001970048309178744,
|
| 1371 |
+
"loss": 1.0324,
|
| 1372 |
"step": 3740
|
| 1373 |
},
|
| 1374 |
{
|
| 1375 |
+
"epoch": 6.539130434782608,
|
| 1376 |
+
"grad_norm": 2.2138895988464355,
|
| 1377 |
+
"learning_rate": 0.00019516908212560387,
|
| 1378 |
+
"loss": 1.0666,
|
| 1379 |
"step": 3760
|
| 1380 |
},
|
| 1381 |
{
|
| 1382 |
+
"epoch": 6.573913043478261,
|
| 1383 |
+
"grad_norm": 1.9186855554580688,
|
| 1384 |
+
"learning_rate": 0.0001932367149758454,
|
| 1385 |
+
"loss": 1.0724,
|
| 1386 |
"step": 3780
|
| 1387 |
},
|
| 1388 |
{
|
| 1389 |
+
"epoch": 6.608695652173913,
|
| 1390 |
+
"grad_norm": 1.302451729774475,
|
| 1391 |
+
"learning_rate": 0.00019159420289855073,
|
| 1392 |
+
"loss": 1.0867,
|
| 1393 |
"step": 3800
|
| 1394 |
},
|
| 1395 |
{
|
| 1396 |
+
"epoch": 6.643478260869565,
|
| 1397 |
+
"grad_norm": 1.1770459413528442,
|
| 1398 |
+
"learning_rate": 0.00018975845410628022,
|
| 1399 |
+
"loss": 1.0659,
|
| 1400 |
"step": 3820
|
| 1401 |
},
|
| 1402 |
{
|
| 1403 |
+
"epoch": 6.678260869565217,
|
| 1404 |
+
"grad_norm": 0.2651650309562683,
|
| 1405 |
+
"learning_rate": 0.0001881159420289855,
|
| 1406 |
+
"loss": 1.0494,
|
| 1407 |
"step": 3840
|
| 1408 |
},
|
| 1409 |
{
|
| 1410 |
+
"epoch": 6.71304347826087,
|
| 1411 |
+
"grad_norm": 0.0,
|
| 1412 |
+
"learning_rate": 0.0001867632850241546,
|
| 1413 |
+
"loss": 1.0464,
|
| 1414 |
"step": 3860
|
| 1415 |
},
|
| 1416 |
{
|
| 1417 |
+
"epoch": 6.747826086956522,
|
| 1418 |
+
"grad_norm": 0.0,
|
| 1419 |
+
"learning_rate": 0.000185024154589372,
|
| 1420 |
+
"loss": 1.0457,
|
| 1421 |
"step": 3880
|
| 1422 |
},
|
| 1423 |
{
|
| 1424 |
+
"epoch": 6.782608695652174,
|
| 1425 |
+
"grad_norm": 0.0,
|
| 1426 |
+
"learning_rate": 0.00018328502415458937,
|
| 1427 |
+
"loss": 0.9815,
|
| 1428 |
"step": 3900
|
| 1429 |
},
|
| 1430 |
{
|
| 1431 |
+
"epoch": 6.817391304347826,
|
| 1432 |
+
"grad_norm": 0.0,
|
| 1433 |
+
"learning_rate": 0.0001816425120772947,
|
| 1434 |
+
"loss": 1.0094,
|
| 1435 |
"step": 3920
|
| 1436 |
},
|
| 1437 |
{
|
| 1438 |
+
"epoch": 6.852173913043478,
|
| 1439 |
+
"grad_norm": NaN,
|
| 1440 |
+
"learning_rate": 0.00018028985507246377,
|
| 1441 |
+
"loss": 1.0023,
|
| 1442 |
"step": 3940
|
| 1443 |
},
|
| 1444 |
{
|
| 1445 |
+
"epoch": 6.886956521739131,
|
| 1446 |
+
"grad_norm": 0.0,
|
| 1447 |
+
"learning_rate": 0.00017893719806763288,
|
| 1448 |
+
"loss": 1.0278,
|
| 1449 |
"step": 3960
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
+
"epoch": 6.921739130434783,
|
| 1453 |
+
"grad_norm": 0.0,
|
| 1454 |
+
"learning_rate": 0.0001771014492753623,
|
| 1455 |
+
"loss": 1.0123,
|
| 1456 |
"step": 3980
|
| 1457 |
},
|
| 1458 |
{
|
| 1459 |
+
"epoch": 6.956521739130435,
|
| 1460 |
+
"grad_norm": 0.0,
|
| 1461 |
+
"learning_rate": 0.00017565217391304346,
|
| 1462 |
+
"loss": 1.0774,
|
| 1463 |
"step": 4000
|
| 1464 |
},
|
| 1465 |
{
|
| 1466 |
+
"epoch": 6.9913043478260875,
|
| 1467 |
+
"grad_norm": 0.0,
|
| 1468 |
+
"learning_rate": 0.00017391304347826088,
|
| 1469 |
+
"loss": 1.0484,
|
| 1470 |
"step": 4020
|
| 1471 |
},
|
| 1472 |
{
|
| 1473 |
+
"epoch": 7.0,
|
| 1474 |
+
"eval_accuracy": 0.9737726967047747,
|
| 1475 |
+
"eval_loss": 0.5780686736106873,
|
| 1476 |
+
"eval_runtime": 118.8154,
|
| 1477 |
+
"eval_samples_per_second": 12.515,
|
| 1478 |
+
"eval_steps_per_second": 12.515,
|
| 1479 |
+
"step": 4025
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 7.026086956521739,
|
| 1483 |
+
"grad_norm": 0.0,
|
| 1484 |
+
"learning_rate": 0.0001723671497584541,
|
| 1485 |
+
"loss": 0.9799,
|
| 1486 |
"step": 4040
|
| 1487 |
},
|
| 1488 |
{
|
| 1489 |
+
"epoch": 7.060869565217391,
|
| 1490 |
+
"grad_norm": 0.0,
|
| 1491 |
+
"learning_rate": 0.00017091787439613525,
|
| 1492 |
+
"loss": 0.9588,
|
| 1493 |
"step": 4060
|
| 1494 |
},
|
| 1495 |
{
|
| 1496 |
+
"epoch": 7.095652173913043,
|
| 1497 |
+
"grad_norm": NaN,
|
| 1498 |
+
"learning_rate": 0.00016966183574879226,
|
| 1499 |
+
"loss": 0.9421,
|
| 1500 |
"step": 4080
|
| 1501 |
},
|
| 1502 |
{
|
| 1503 |
+
"epoch": 7.130434782608695,
|
| 1504 |
+
"grad_norm": 0.0,
|
| 1505 |
+
"learning_rate": 0.00016782608695652175,
|
| 1506 |
+
"loss": 0.9551,
|
| 1507 |
"step": 4100
|
| 1508 |
},
|
| 1509 |
{
|
| 1510 |
+
"epoch": 7.165217391304348,
|
| 1511 |
+
"grad_norm": 0.0,
|
| 1512 |
+
"learning_rate": 0.00016618357487922704,
|
| 1513 |
+
"loss": 0.9622,
|
| 1514 |
"step": 4120
|
| 1515 |
},
|
| 1516 |
{
|
| 1517 |
+
"epoch": 7.2,
|
| 1518 |
+
"grad_norm": 0.0,
|
| 1519 |
+
"learning_rate": 0.00016444444444444446,
|
| 1520 |
+
"loss": 0.9712,
|
| 1521 |
"step": 4140
|
| 1522 |
},
|
| 1523 |
{
|
| 1524 |
+
"epoch": 7.234782608695652,
|
| 1525 |
+
"grad_norm": 0.0,
|
| 1526 |
+
"learning_rate": 0.00016299516908212561,
|
| 1527 |
+
"loss": 0.9834,
|
| 1528 |
"step": 4160
|
| 1529 |
},
|
| 1530 |
{
|
| 1531 |
+
"epoch": 7.269565217391304,
|
| 1532 |
+
"grad_norm": NaN,
|
| 1533 |
+
"learning_rate": 0.00016135265700483093,
|
| 1534 |
+
"loss": 0.9968,
|
| 1535 |
"step": 4180
|
| 1536 |
},
|
| 1537 |
{
|
| 1538 |
+
"epoch": 7.304347826086957,
|
| 1539 |
+
"grad_norm": 0.0,
|
| 1540 |
+
"learning_rate": 0.00015961352657004833,
|
| 1541 |
+
"loss": 0.956,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1542 |
"step": 4200
|
| 1543 |
},
|
| 1544 |
{
|
| 1545 |
+
"epoch": 7.339130434782609,
|
| 1546 |
+
"grad_norm": 0.0,
|
| 1547 |
+
"learning_rate": 0.00015806763285024155,
|
| 1548 |
+
"loss": 0.8981,
|
| 1549 |
"step": 4220
|
| 1550 |
},
|
| 1551 |
{
|
| 1552 |
+
"epoch": 7.373913043478261,
|
| 1553 |
+
"grad_norm": 0.0,
|
| 1554 |
+
"learning_rate": 0.00015642512077294684,
|
| 1555 |
+
"loss": 0.9515,
|
| 1556 |
"step": 4240
|
| 1557 |
},
|
| 1558 |
{
|
| 1559 |
+
"epoch": 7.408695652173913,
|
| 1560 |
+
"grad_norm": 0.0,
|
| 1561 |
+
"learning_rate": 0.0001548792270531401,
|
| 1562 |
+
"loss": 0.9535,
|
| 1563 |
"step": 4260
|
| 1564 |
},
|
| 1565 |
{
|
| 1566 |
+
"epoch": 7.443478260869565,
|
| 1567 |
+
"grad_norm": NaN,
|
| 1568 |
+
"learning_rate": 0.00015333333333333334,
|
| 1569 |
+
"loss": 0.9646,
|
| 1570 |
"step": 4280
|
| 1571 |
},
|
| 1572 |
{
|
| 1573 |
+
"epoch": 7.478260869565218,
|
| 1574 |
+
"grad_norm": 0.0,
|
| 1575 |
+
"learning_rate": 0.00015140096618357487,
|
| 1576 |
+
"loss": 0.9821,
|
| 1577 |
"step": 4300
|
| 1578 |
},
|
| 1579 |
{
|
| 1580 |
+
"epoch": 7.51304347826087,
|
| 1581 |
+
"grad_norm": 0.0,
|
| 1582 |
+
"learning_rate": 0.00015014492753623188,
|
| 1583 |
+
"loss": 0.9259,
|
| 1584 |
"step": 4320
|
| 1585 |
},
|
| 1586 |
{
|
| 1587 |
+
"epoch": 7.547826086956522,
|
| 1588 |
+
"grad_norm": 0.0,
|
| 1589 |
+
"learning_rate": 0.00014869565217391303,
|
| 1590 |
+
"loss": 0.9494,
|
| 1591 |
"step": 4340
|
| 1592 |
},
|
| 1593 |
{
|
| 1594 |
+
"epoch": 7.582608695652174,
|
| 1595 |
+
"grad_norm": 0.0,
|
| 1596 |
+
"learning_rate": 0.00014714975845410628,
|
| 1597 |
+
"loss": 0.9305,
|
| 1598 |
"step": 4360
|
| 1599 |
},
|
| 1600 |
{
|
| 1601 |
+
"epoch": 7.6173913043478265,
|
| 1602 |
+
"grad_norm": 0.0,
|
| 1603 |
+
"learning_rate": 0.0001455072463768116,
|
| 1604 |
+
"loss": 0.8889,
|
| 1605 |
"step": 4380
|
| 1606 |
},
|
| 1607 |
{
|
| 1608 |
+
"epoch": 7.6521739130434785,
|
| 1609 |
+
"grad_norm": 0.0,
|
| 1610 |
+
"learning_rate": 0.00014396135265700482,
|
| 1611 |
+
"loss": 0.9524,
|
| 1612 |
"step": 4400
|
| 1613 |
},
|
| 1614 |
{
|
| 1615 |
+
"epoch": 7.6869565217391305,
|
| 1616 |
+
"grad_norm": 0.0,
|
| 1617 |
+
"learning_rate": 0.00014231884057971014,
|
| 1618 |
+
"loss": 0.9065,
|
| 1619 |
"step": 4420
|
| 1620 |
},
|
| 1621 |
{
|
| 1622 |
+
"epoch": 7.721739130434782,
|
| 1623 |
+
"grad_norm": 0.0,
|
| 1624 |
+
"learning_rate": 0.00014048309178743963,
|
| 1625 |
+
"loss": 0.9153,
|
| 1626 |
"step": 4440
|
| 1627 |
},
|
| 1628 |
{
|
| 1629 |
+
"epoch": 7.756521739130434,
|
| 1630 |
+
"grad_norm": NaN,
|
| 1631 |
+
"learning_rate": 0.0001403864734299517,
|
| 1632 |
+
"loss": 0.6675,
|
| 1633 |
"step": 4460
|
| 1634 |
},
|
| 1635 |
{
|
| 1636 |
+
"epoch": 7.791304347826087,
|
| 1637 |
+
"grad_norm": NaN,
|
| 1638 |
+
"learning_rate": 0.0001403864734299517,
|
| 1639 |
+
"loss": 0.0,
|
| 1640 |
"step": 4480
|
| 1641 |
},
|
| 1642 |
{
|
| 1643 |
+
"epoch": 7.826086956521739,
|
| 1644 |
+
"grad_norm": NaN,
|
| 1645 |
+
"learning_rate": 0.0001403864734299517,
|
| 1646 |
+
"loss": 0.0,
|
| 1647 |
"step": 4500
|
| 1648 |
},
|
| 1649 |
{
|
| 1650 |
+
"epoch": 7.860869565217391,
|
| 1651 |
+
"grad_norm": NaN,
|
| 1652 |
+
"learning_rate": 0.0001403864734299517,
|
| 1653 |
+
"loss": 0.0,
|
| 1654 |
"step": 4520
|
| 1655 |
},
|
| 1656 |
{
|
| 1657 |
+
"epoch": 7.895652173913043,
|
| 1658 |
+
"grad_norm": NaN,
|
| 1659 |
+
"learning_rate": 0.0001403864734299517,
|
| 1660 |
+
"loss": 0.0,
|
| 1661 |
"step": 4540
|
| 1662 |
},
|
| 1663 |
{
|
| 1664 |
+
"epoch": 7.930434782608696,
|
| 1665 |
+
"grad_norm": NaN,
|
| 1666 |
+
"learning_rate": 0.0001403864734299517,
|
| 1667 |
+
"loss": 0.0,
|
| 1668 |
"step": 4560
|
| 1669 |
},
|
| 1670 |
{
|
| 1671 |
+
"epoch": 7.965217391304348,
|
| 1672 |
+
"grad_norm": NaN,
|
| 1673 |
+
"learning_rate": 0.0001403864734299517,
|
| 1674 |
+
"loss": 0.0,
|
| 1675 |
"step": 4580
|
| 1676 |
},
|
| 1677 |
{
|
| 1678 |
+
"epoch": 8.0,
|
| 1679 |
+
"grad_norm": NaN,
|
| 1680 |
+
"learning_rate": 0.0001403864734299517,
|
| 1681 |
+
"loss": 0.0,
|
| 1682 |
+
"step": 4600
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 8.0,
|
| 1686 |
+
"eval_accuracy": 0.0006724949562878278,
|
| 1687 |
+
"eval_loss": NaN,
|
| 1688 |
+
"eval_runtime": 129.6238,
|
| 1689 |
+
"eval_samples_per_second": 11.472,
|
| 1690 |
+
"eval_steps_per_second": 11.472,
|
| 1691 |
"step": 4600
|
| 1692 |
},
|
| 1693 |
{
|
| 1694 |
+
"epoch": 8.034782608695652,
|
| 1695 |
+
"grad_norm": NaN,
|
| 1696 |
+
"learning_rate": 0.0001403864734299517,
|
| 1697 |
+
"loss": 0.0,
|
| 1698 |
"step": 4620
|
| 1699 |
},
|
| 1700 |
{
|
| 1701 |
+
"epoch": 8.069565217391304,
|
| 1702 |
+
"grad_norm": NaN,
|
| 1703 |
+
"learning_rate": 0.0001403864734299517,
|
| 1704 |
+
"loss": 0.0,
|
| 1705 |
"step": 4640
|
| 1706 |
},
|
| 1707 |
{
|
| 1708 |
+
"epoch": 8.104347826086956,
|
| 1709 |
+
"grad_norm": NaN,
|
| 1710 |
+
"learning_rate": 0.0001403864734299517,
|
| 1711 |
+
"loss": 0.0,
|
| 1712 |
"step": 4660
|
| 1713 |
},
|
| 1714 |
{
|
| 1715 |
+
"epoch": 8.139130434782608,
|
| 1716 |
+
"grad_norm": NaN,
|
| 1717 |
+
"learning_rate": 0.0001403864734299517,
|
| 1718 |
+
"loss": 0.0,
|
| 1719 |
"step": 4680
|
| 1720 |
},
|
| 1721 |
{
|
| 1722 |
+
"epoch": 8.173913043478262,
|
| 1723 |
+
"grad_norm": NaN,
|
| 1724 |
+
"learning_rate": 0.0001403864734299517,
|
| 1725 |
+
"loss": 0.0,
|
| 1726 |
"step": 4700
|
| 1727 |
},
|
| 1728 |
{
|
| 1729 |
+
"epoch": 8.208695652173914,
|
| 1730 |
+
"grad_norm": NaN,
|
| 1731 |
+
"learning_rate": 0.0001403864734299517,
|
| 1732 |
+
"loss": 0.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1733 |
"step": 4720
|
| 1734 |
},
|
| 1735 |
{
|
| 1736 |
+
"epoch": 8.243478260869566,
|
| 1737 |
+
"grad_norm": NaN,
|
| 1738 |
+
"learning_rate": 0.0001403864734299517,
|
| 1739 |
+
"loss": 0.0,
|
| 1740 |
"step": 4740
|
| 1741 |
},
|
| 1742 |
{
|
| 1743 |
+
"epoch": 8.278260869565218,
|
| 1744 |
+
"grad_norm": NaN,
|
| 1745 |
+
"learning_rate": 0.0001403864734299517,
|
| 1746 |
+
"loss": 0.0,
|
| 1747 |
"step": 4760
|
| 1748 |
},
|
| 1749 |
{
|
| 1750 |
+
"epoch": 8.31304347826087,
|
| 1751 |
+
"grad_norm": NaN,
|
| 1752 |
+
"learning_rate": 0.0001403864734299517,
|
| 1753 |
+
"loss": 0.0,
|
| 1754 |
"step": 4780
|
| 1755 |
},
|
| 1756 |
{
|
| 1757 |
+
"epoch": 8.347826086956522,
|
| 1758 |
+
"grad_norm": NaN,
|
| 1759 |
+
"learning_rate": 0.0001403864734299517,
|
| 1760 |
+
"loss": 0.0,
|
| 1761 |
"step": 4800
|
| 1762 |
},
|
| 1763 |
{
|
| 1764 |
+
"epoch": 8.382608695652173,
|
| 1765 |
+
"grad_norm": NaN,
|
| 1766 |
+
"learning_rate": 0.0001403864734299517,
|
| 1767 |
+
"loss": 0.0,
|
| 1768 |
"step": 4820
|
| 1769 |
},
|
| 1770 |
{
|
| 1771 |
+
"epoch": 8.417391304347825,
|
| 1772 |
+
"grad_norm": NaN,
|
| 1773 |
+
"learning_rate": 0.0001403864734299517,
|
| 1774 |
+
"loss": 0.0,
|
| 1775 |
"step": 4840
|
| 1776 |
},
|
| 1777 |
{
|
| 1778 |
+
"epoch": 8.452173913043477,
|
| 1779 |
+
"grad_norm": NaN,
|
| 1780 |
+
"learning_rate": 0.0001403864734299517,
|
| 1781 |
+
"loss": 0.0,
|
| 1782 |
"step": 4860
|
| 1783 |
},
|
| 1784 |
{
|
| 1785 |
+
"epoch": 8.486956521739131,
|
| 1786 |
+
"grad_norm": NaN,
|
| 1787 |
+
"learning_rate": 0.0001403864734299517,
|
| 1788 |
+
"loss": 0.0,
|
| 1789 |
"step": 4880
|
| 1790 |
},
|
| 1791 |
{
|
| 1792 |
+
"epoch": 8.521739130434783,
|
| 1793 |
+
"grad_norm": NaN,
|
| 1794 |
+
"learning_rate": 0.0001403864734299517,
|
| 1795 |
+
"loss": 0.0,
|
| 1796 |
"step": 4900
|
| 1797 |
},
|
| 1798 |
{
|
| 1799 |
+
"epoch": 8.556521739130435,
|
| 1800 |
+
"grad_norm": NaN,
|
| 1801 |
+
"learning_rate": 0.0001403864734299517,
|
| 1802 |
+
"loss": 0.0,
|
| 1803 |
"step": 4920
|
| 1804 |
},
|
| 1805 |
{
|
| 1806 |
+
"epoch": 8.591304347826087,
|
| 1807 |
+
"grad_norm": NaN,
|
| 1808 |
+
"learning_rate": 0.0001403864734299517,
|
| 1809 |
+
"loss": 0.0,
|
| 1810 |
"step": 4940
|
| 1811 |
},
|
| 1812 |
{
|
| 1813 |
+
"epoch": 8.626086956521739,
|
| 1814 |
+
"grad_norm": NaN,
|
| 1815 |
+
"learning_rate": 0.0001403864734299517,
|
| 1816 |
+
"loss": 0.0,
|
| 1817 |
"step": 4960
|
| 1818 |
},
|
| 1819 |
{
|
| 1820 |
+
"epoch": 8.660869565217391,
|
| 1821 |
+
"grad_norm": NaN,
|
| 1822 |
+
"learning_rate": 0.0001403864734299517,
|
| 1823 |
+
"loss": 0.0,
|
| 1824 |
"step": 4980
|
| 1825 |
},
|
| 1826 |
{
|
| 1827 |
+
"epoch": 8.695652173913043,
|
| 1828 |
+
"grad_norm": NaN,
|
| 1829 |
+
"learning_rate": 0.0001403864734299517,
|
| 1830 |
+
"loss": 0.0,
|
| 1831 |
"step": 5000
|
| 1832 |
},
|
| 1833 |
{
|
| 1834 |
+
"epoch": 8.730434782608695,
|
| 1835 |
+
"grad_norm": NaN,
|
| 1836 |
+
"learning_rate": 0.0001403864734299517,
|
| 1837 |
+
"loss": 0.0,
|
| 1838 |
"step": 5020
|
| 1839 |
},
|
| 1840 |
{
|
| 1841 |
+
"epoch": 8.765217391304347,
|
| 1842 |
+
"grad_norm": NaN,
|
| 1843 |
+
"learning_rate": 0.0001403864734299517,
|
| 1844 |
+
"loss": 0.0,
|
| 1845 |
"step": 5040
|
| 1846 |
},
|
| 1847 |
{
|
| 1848 |
+
"epoch": 8.8,
|
| 1849 |
+
"grad_norm": NaN,
|
| 1850 |
+
"learning_rate": 0.0001403864734299517,
|
| 1851 |
+
"loss": 0.0,
|
| 1852 |
"step": 5060
|
| 1853 |
},
|
| 1854 |
{
|
| 1855 |
+
"epoch": 8.834782608695653,
|
| 1856 |
+
"grad_norm": NaN,
|
| 1857 |
+
"learning_rate": 0.0001403864734299517,
|
| 1858 |
+
"loss": 0.0,
|
| 1859 |
"step": 5080
|
| 1860 |
},
|
| 1861 |
{
|
| 1862 |
+
"epoch": 8.869565217391305,
|
| 1863 |
+
"grad_norm": NaN,
|
| 1864 |
+
"learning_rate": 0.0001403864734299517,
|
| 1865 |
+
"loss": 0.0,
|
| 1866 |
"step": 5100
|
| 1867 |
},
|
| 1868 |
{
|
| 1869 |
+
"epoch": 8.904347826086957,
|
| 1870 |
+
"grad_norm": NaN,
|
| 1871 |
+
"learning_rate": 0.0001403864734299517,
|
| 1872 |
+
"loss": 0.0,
|
| 1873 |
"step": 5120
|
| 1874 |
},
|
| 1875 |
{
|
| 1876 |
+
"epoch": 8.939130434782609,
|
| 1877 |
+
"grad_norm": NaN,
|
| 1878 |
+
"learning_rate": 0.0001403864734299517,
|
| 1879 |
+
"loss": 0.0,
|
| 1880 |
"step": 5140
|
| 1881 |
},
|
| 1882 |
{
|
| 1883 |
+
"epoch": 8.97391304347826,
|
| 1884 |
+
"grad_norm": NaN,
|
| 1885 |
+
"learning_rate": 0.0001403864734299517,
|
| 1886 |
+
"loss": 0.0,
|
| 1887 |
"step": 5160
|
| 1888 |
},
|
| 1889 |
{
|
| 1890 |
+
"epoch": 9.0,
|
| 1891 |
+
"eval_accuracy": 0.0006724949562878278,
|
| 1892 |
+
"eval_loss": NaN,
|
| 1893 |
+
"eval_runtime": 117.1288,
|
| 1894 |
+
"eval_samples_per_second": 12.695,
|
| 1895 |
+
"eval_steps_per_second": 12.695,
|
| 1896 |
+
"step": 5175
|
| 1897 |
+
},
|
| 1898 |
+
{
|
| 1899 |
+
"epoch": 9.008695652173913,
|
| 1900 |
+
"grad_norm": NaN,
|
| 1901 |
+
"learning_rate": 0.0001403864734299517,
|
| 1902 |
+
"loss": 0.0,
|
| 1903 |
"step": 5180
|
| 1904 |
},
|
| 1905 |
{
|
| 1906 |
+
"epoch": 9.043478260869565,
|
| 1907 |
+
"grad_norm": NaN,
|
| 1908 |
+
"learning_rate": 0.0001403864734299517,
|
| 1909 |
+
"loss": 0.0,
|
| 1910 |
"step": 5200
|
| 1911 |
},
|
| 1912 |
{
|
| 1913 |
+
"epoch": 9.078260869565218,
|
| 1914 |
+
"grad_norm": NaN,
|
| 1915 |
+
"learning_rate": 0.0001403864734299517,
|
| 1916 |
+
"loss": 0.0,
|
| 1917 |
"step": 5220
|
| 1918 |
},
|
| 1919 |
+
{
|
| 1920 |
+
"epoch": 9.11304347826087,
|
| 1921 |
+
"grad_norm": NaN,
|
| 1922 |
+
"learning_rate": 0.0001403864734299517,
|
| 1923 |
+
"loss": 0.0,
|
| 1924 |
+
"step": 5240
|
| 1925 |
+
},
|
| 1926 |
+
{
|
| 1927 |
+
"epoch": 9.147826086956522,
|
| 1928 |
+
"grad_norm": NaN,
|
| 1929 |
+
"learning_rate": 0.0001403864734299517,
|
| 1930 |
+
"loss": 0.0,
|
| 1931 |
+
"step": 5260
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 9.182608695652174,
|
| 1935 |
+
"grad_norm": NaN,
|
| 1936 |
+
"learning_rate": 0.0001403864734299517,
|
| 1937 |
+
"loss": 0.0,
|
| 1938 |
+
"step": 5280
|
| 1939 |
+
},
|
| 1940 |
+
{
|
| 1941 |
+
"epoch": 9.217391304347826,
|
| 1942 |
+
"grad_norm": NaN,
|
| 1943 |
+
"learning_rate": 0.0001403864734299517,
|
| 1944 |
+
"loss": 0.0,
|
| 1945 |
+
"step": 5300
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 9.252173913043478,
|
| 1949 |
+
"grad_norm": NaN,
|
| 1950 |
+
"learning_rate": 0.0001403864734299517,
|
| 1951 |
+
"loss": 0.0,
|
| 1952 |
+
"step": 5320
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 9.28695652173913,
|
| 1956 |
+
"grad_norm": NaN,
|
| 1957 |
+
"learning_rate": 0.0001403864734299517,
|
| 1958 |
+
"loss": 0.0,
|
| 1959 |
+
"step": 5340
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 9.321739130434782,
|
| 1963 |
+
"grad_norm": NaN,
|
| 1964 |
+
"learning_rate": 0.0001403864734299517,
|
| 1965 |
+
"loss": 0.0,
|
| 1966 |
+
"step": 5360
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 9.356521739130434,
|
| 1970 |
+
"grad_norm": NaN,
|
| 1971 |
+
"learning_rate": 0.0001403864734299517,
|
| 1972 |
+
"loss": 0.0,
|
| 1973 |
+
"step": 5380
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 9.391304347826088,
|
| 1977 |
+
"grad_norm": NaN,
|
| 1978 |
+
"learning_rate": 0.0001403864734299517,
|
| 1979 |
+
"loss": 0.0,
|
| 1980 |
+
"step": 5400
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 9.42608695652174,
|
| 1984 |
+
"grad_norm": NaN,
|
| 1985 |
+
"learning_rate": 0.0001403864734299517,
|
| 1986 |
+
"loss": 0.0,
|
| 1987 |
+
"step": 5420
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 9.460869565217392,
|
| 1991 |
+
"grad_norm": NaN,
|
| 1992 |
+
"learning_rate": 0.0001403864734299517,
|
| 1993 |
+
"loss": 0.0,
|
| 1994 |
+
"step": 5440
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 9.495652173913044,
|
| 1998 |
+
"grad_norm": NaN,
|
| 1999 |
+
"learning_rate": 0.0001403864734299517,
|
| 2000 |
+
"loss": 0.0,
|
| 2001 |
+
"step": 5460
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 9.530434782608696,
|
| 2005 |
+
"grad_norm": NaN,
|
| 2006 |
+
"learning_rate": 0.0001403864734299517,
|
| 2007 |
+
"loss": 0.0,
|
| 2008 |
+
"step": 5480
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 9.565217391304348,
|
| 2012 |
+
"grad_norm": NaN,
|
| 2013 |
+
"learning_rate": 0.0001403864734299517,
|
| 2014 |
+
"loss": 0.0,
|
| 2015 |
+
"step": 5500
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 9.6,
|
| 2019 |
+
"grad_norm": NaN,
|
| 2020 |
+
"learning_rate": 0.0001403864734299517,
|
| 2021 |
+
"loss": 0.0,
|
| 2022 |
+
"step": 5520
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 9.634782608695652,
|
| 2026 |
+
"grad_norm": NaN,
|
| 2027 |
+
"learning_rate": 0.0001403864734299517,
|
| 2028 |
+
"loss": 0.0,
|
| 2029 |
+
"step": 5540
|
| 2030 |
+
},
|
| 2031 |
+
{
|
| 2032 |
+
"epoch": 9.669565217391304,
|
| 2033 |
+
"grad_norm": NaN,
|
| 2034 |
+
"learning_rate": 0.0001403864734299517,
|
| 2035 |
+
"loss": 0.0,
|
| 2036 |
+
"step": 5560
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 9.704347826086957,
|
| 2040 |
+
"grad_norm": NaN,
|
| 2041 |
+
"learning_rate": 0.0001403864734299517,
|
| 2042 |
+
"loss": 0.0,
|
| 2043 |
+
"step": 5580
|
| 2044 |
+
},
|
| 2045 |
+
{
|
| 2046 |
+
"epoch": 9.73913043478261,
|
| 2047 |
+
"grad_norm": NaN,
|
| 2048 |
+
"learning_rate": 0.0001403864734299517,
|
| 2049 |
+
"loss": 0.0,
|
| 2050 |
+
"step": 5600
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 9.773913043478261,
|
| 2054 |
+
"grad_norm": NaN,
|
| 2055 |
+
"learning_rate": 0.0001403864734299517,
|
| 2056 |
+
"loss": 0.0,
|
| 2057 |
+
"step": 5620
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 9.808695652173913,
|
| 2061 |
+
"grad_norm": NaN,
|
| 2062 |
+
"learning_rate": 0.0001403864734299517,
|
| 2063 |
+
"loss": 0.0,
|
| 2064 |
+
"step": 5640
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 9.843478260869565,
|
| 2068 |
+
"grad_norm": NaN,
|
| 2069 |
+
"learning_rate": 0.0001403864734299517,
|
| 2070 |
+
"loss": 0.0,
|
| 2071 |
+
"step": 5660
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 9.878260869565217,
|
| 2075 |
+
"grad_norm": NaN,
|
| 2076 |
+
"learning_rate": 0.0001403864734299517,
|
| 2077 |
+
"loss": 0.0,
|
| 2078 |
+
"step": 5680
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 9.91304347826087,
|
| 2082 |
+
"grad_norm": NaN,
|
| 2083 |
+
"learning_rate": 0.0001403864734299517,
|
| 2084 |
+
"loss": 0.0,
|
| 2085 |
+
"step": 5700
|
| 2086 |
+
},
|
| 2087 |
+
{
|
| 2088 |
+
"epoch": 9.947826086956521,
|
| 2089 |
+
"grad_norm": NaN,
|
| 2090 |
+
"learning_rate": 0.0001403864734299517,
|
| 2091 |
+
"loss": 0.0,
|
| 2092 |
+
"step": 5720
|
| 2093 |
+
},
|
| 2094 |
+
{
|
| 2095 |
+
"epoch": 9.982608695652173,
|
| 2096 |
+
"grad_norm": NaN,
|
| 2097 |
+
"learning_rate": 0.0001403864734299517,
|
| 2098 |
+
"loss": 0.0,
|
| 2099 |
+
"step": 5740
|
| 2100 |
+
},
|
| 2101 |
{
|
| 2102 |
"epoch": 10.0,
|
| 2103 |
+
"eval_accuracy": 0.0006724949562878278,
|
| 2104 |
+
"eval_loss": NaN,
|
| 2105 |
+
"eval_runtime": 103.3199,
|
| 2106 |
+
"eval_samples_per_second": 14.392,
|
| 2107 |
+
"eval_steps_per_second": 14.392,
|
| 2108 |
+
"step": 5750
|
| 2109 |
},
|
| 2110 |
{
|
| 2111 |
"epoch": 10.0,
|
| 2112 |
+
"step": 5750,
|
| 2113 |
+
"total_flos": 2.7398100529152e+18,
|
| 2114 |
+
"train_loss": 2.9414075751926587,
|
| 2115 |
+
"train_runtime": 59857.6179,
|
| 2116 |
+
"train_samples_per_second": 24.584,
|
| 2117 |
+
"train_steps_per_second": 0.096
|
| 2118 |
}
|
| 2119 |
],
|
| 2120 |
"logging_steps": 20,
|
| 2121 |
+
"max_steps": 5750,
|
| 2122 |
"num_input_tokens_seen": 0,
|
| 2123 |
"num_train_epochs": 10,
|
| 2124 |
"save_steps": 500,
|
|
|
|
| 2134 |
"attributes": {}
|
| 2135 |
}
|
| 2136 |
},
|
| 2137 |
+
"total_flos": 2.7398100529152e+18,
|
| 2138 |
"train_batch_size": 256,
|
| 2139 |
"trial_name": null,
|
| 2140 |
"trial_params": null
|