End of training
Browse files- README.md +25 -19
- pytorch_model.bin +1 -1
- trainer_state.json +932 -512
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -15,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [alexue4/text-normalization-ru-new](https://huggingface.co/alexue4/text-normalization-ru-new) on the None dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
- Mean Distance: 0
|
| 20 |
-
- Max Distance:
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
@@ -43,26 +43,32 @@ The following hyperparameters were used during training:
|
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
- lr_scheduler_warmup_ratio: 0.1
|
| 46 |
-
- num_epochs:
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
-
|
| 52 |
-
| 0.
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
|
| 67 |
|
| 68 |
### Framework versions
|
|
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [alexue4/text-normalization-ru-new](https://huggingface.co/alexue4/text-normalization-ru-new) on the None dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.0052
|
| 19 |
- Mean Distance: 0
|
| 20 |
+
- Max Distance: 1
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
- lr_scheduler_warmup_ratio: 0.1
|
| 46 |
+
- num_epochs: 20
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss | Mean Distance | Max Distance |
|
| 51 |
+
|:-------------:|:-----:|:------:|:---------------:|:-------------:|:------------:|
|
| 52 |
+
| 0.0011 | 1.0 | 23139 | 0.0060 | 0 | 4 |
|
| 53 |
+
| 0.0013 | 2.0 | 46278 | 0.0038 | 0 | 4 |
|
| 54 |
+
| 0.0014 | 3.0 | 69417 | 0.0036 | 0 | 4 |
|
| 55 |
+
| 0.0014 | 4.0 | 92556 | 0.0044 | 0 | 4 |
|
| 56 |
+
| 0.0013 | 5.0 | 115695 | 0.0056 | 0 | 4 |
|
| 57 |
+
| 0.0014 | 6.0 | 138834 | 0.0034 | 0 | 2 |
|
| 58 |
+
| 0.0012 | 7.0 | 161973 | 0.0036 | 0 | 2 |
|
| 59 |
+
| 0.0011 | 8.0 | 185112 | 0.0025 | 0 | 1 |
|
| 60 |
+
| 0.0011 | 9.0 | 208251 | 0.0041 | 0 | 4 |
|
| 61 |
+
| 0.001 | 10.0 | 231390 | 0.0064 | 0 | 1 |
|
| 62 |
+
| 0.001 | 11.0 | 254529 | 0.0059 | 0 | 1 |
|
| 63 |
+
| 0.0008 | 12.0 | 277668 | 0.0046 | 0 | 3 |
|
| 64 |
+
| 0.0008 | 13.0 | 300807 | 0.0041 | 0 | 2 |
|
| 65 |
+
| 0.0008 | 14.0 | 323946 | 0.0059 | 0 | 2 |
|
| 66 |
+
| 0.0008 | 15.0 | 347085 | 0.0060 | 0 | 2 |
|
| 67 |
+
| 0.0007 | 16.0 | 370224 | 0.0053 | 0 | 2 |
|
| 68 |
+
| 0.0007 | 17.0 | 393363 | 0.0041 | 0 | 1 |
|
| 69 |
+
| 0.0007 | 18.0 | 416502 | 0.0049 | 0 | 1 |
|
| 70 |
+
| 0.0007 | 19.0 | 439641 | 0.0054 | 0 | 1 |
|
| 71 |
+
| 0.0006 | 20.0 | 462780 | 0.0052 | 0 | 1 |
|
| 72 |
|
| 73 |
|
| 74 |
### Framework versions
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 258643461
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:578de3092572f82fad85802a1bc63b97248c2ea001457ea87093eca100ca8078
|
| 3 |
size 258643461
|
trainer_state.json
CHANGED
|
@@ -1,1008 +1,1428 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 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": 0.0,
|
| 13 |
-
"learning_rate": 2.
|
| 14 |
-
"loss": 0.
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.1,
|
| 19 |
-
"learning_rate": 5.
|
| 20 |
-
"loss": 0.
|
| 21 |
-
"step":
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.2,
|
| 25 |
-
"learning_rate": 1.
|
| 26 |
-
"loss": 0.
|
| 27 |
-
"step":
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.3,
|
| 31 |
-
"learning_rate": 1.
|
| 32 |
-
"loss": 0.
|
| 33 |
-
"step":
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.4,
|
| 37 |
-
"learning_rate": 2.
|
| 38 |
-
"loss": 0.
|
| 39 |
-
"step":
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.5,
|
| 43 |
-
"learning_rate": 2.
|
| 44 |
-
"loss": 0.
|
| 45 |
-
"step":
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.6,
|
| 49 |
-
"learning_rate": 3.
|
| 50 |
-
"loss": 0.
|
| 51 |
-
"step":
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.7,
|
| 55 |
-
"learning_rate": 3.
|
| 56 |
-
"loss": 0.
|
| 57 |
-
"step":
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.8,
|
| 61 |
-
"learning_rate": 4.
|
| 62 |
-
"loss": 0.
|
| 63 |
-
"step":
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.9,
|
| 67 |
-
"learning_rate": 4.
|
| 68 |
-
"loss": 0.
|
| 69 |
-
"step":
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 1.0,
|
| 73 |
-
"eval_loss": 0.
|
| 74 |
-
"eval_max_distance":
|
| 75 |
"eval_mean_distance": 0,
|
| 76 |
-
"eval_runtime":
|
| 77 |
-
"eval_samples_per_second":
|
| 78 |
-
"eval_steps_per_second":
|
| 79 |
-
"step":
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.0,
|
| 83 |
-
"learning_rate": 5.
|
| 84 |
-
"loss": 0.
|
| 85 |
-
"step":
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"epoch": 1.1,
|
| 89 |
-
"learning_rate": 5.
|
| 90 |
-
"loss": 0.
|
| 91 |
-
"step":
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 1.2,
|
| 95 |
-
"learning_rate": 6.
|
| 96 |
-
"loss": 0.
|
| 97 |
-
"step":
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 1.3,
|
| 101 |
-
"learning_rate": 6.
|
| 102 |
-
"loss": 0.
|
| 103 |
-
"step":
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 1.4,
|
| 107 |
-
"learning_rate": 7.
|
| 108 |
-
"loss": 0.
|
| 109 |
-
"step":
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 1.5,
|
| 113 |
-
"learning_rate": 7.
|
| 114 |
-
"loss": 0.
|
| 115 |
-
"step":
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 1.6,
|
| 119 |
-
"learning_rate": 8.
|
| 120 |
-
"loss": 0.
|
| 121 |
-
"step":
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 1.7,
|
| 125 |
-
"learning_rate": 8.
|
| 126 |
-
"loss": 0.
|
| 127 |
-
"step":
|
| 128 |
},
|
| 129 |
{
|
| 130 |
"epoch": 1.8,
|
| 131 |
-
"learning_rate": 9.
|
| 132 |
-
"loss": 0.
|
| 133 |
-
"step":
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 1.9,
|
| 137 |
-
"learning_rate": 9.
|
| 138 |
-
"loss": 0.
|
| 139 |
-
"step":
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 2.0,
|
| 143 |
-
"eval_loss": 0.
|
| 144 |
-
"eval_max_distance":
|
| 145 |
"eval_mean_distance": 0,
|
| 146 |
-
"eval_runtime":
|
| 147 |
-
"eval_samples_per_second":
|
| 148 |
-
"eval_steps_per_second":
|
| 149 |
-
"step":
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 2.0,
|
| 153 |
-
"learning_rate": 9.
|
| 154 |
-
"loss": 0.
|
| 155 |
-
"step":
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 2.1,
|
| 159 |
-
"learning_rate": 9.
|
| 160 |
-
"loss": 0.
|
| 161 |
-
"step":
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 2.2,
|
| 165 |
-
"learning_rate": 9.
|
| 166 |
-
"loss": 0.
|
| 167 |
-
"step":
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 2.3,
|
| 171 |
-
"learning_rate": 9.
|
| 172 |
-
"loss": 0.
|
| 173 |
-
"step":
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 2.4,
|
| 177 |
-
"learning_rate": 9.
|
| 178 |
-
"loss": 0.
|
| 179 |
-
"step":
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 2.5,
|
| 183 |
-
"learning_rate": 9.
|
| 184 |
-
"loss": 0.
|
| 185 |
-
"step":
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 2.6,
|
| 189 |
-
"learning_rate": 9.
|
| 190 |
-
"loss": 0.
|
| 191 |
-
"step":
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 2.7,
|
| 195 |
-
"learning_rate": 9.
|
| 196 |
-
"loss": 0.
|
| 197 |
-
"step":
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 2.8,
|
| 201 |
-
"learning_rate": 9.
|
| 202 |
-
"loss": 0.
|
| 203 |
-
"step":
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 2.9,
|
| 207 |
-
"learning_rate": 9.
|
| 208 |
-
"loss": 0.
|
| 209 |
-
"step":
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 3.0,
|
| 213 |
-
"eval_loss": 0.
|
| 214 |
-
"eval_max_distance":
|
| 215 |
"eval_mean_distance": 0,
|
| 216 |
-
"eval_runtime":
|
| 217 |
-
"eval_samples_per_second":
|
| 218 |
-
"eval_steps_per_second": 1.
|
| 219 |
-
"step":
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 3.0,
|
| 223 |
-
"learning_rate": 9.
|
| 224 |
-
"loss": 0.
|
| 225 |
-
"step":
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 3.1,
|
| 229 |
-
"learning_rate": 9.
|
| 230 |
-
"loss": 0.
|
| 231 |
-
"step":
|
| 232 |
},
|
| 233 |
{
|
| 234 |
"epoch": 3.2,
|
| 235 |
-
"learning_rate": 9.
|
| 236 |
-
"loss": 0.
|
| 237 |
-
"step":
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 3.3,
|
| 241 |
-
"learning_rate": 9.
|
| 242 |
-
"loss": 0.
|
| 243 |
-
"step":
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 3.4,
|
| 247 |
-
"learning_rate": 9.
|
| 248 |
-
"loss": 0.
|
| 249 |
-
"step":
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 3.5,
|
| 253 |
-
"learning_rate": 9.
|
| 254 |
-
"loss": 0.
|
| 255 |
-
"step":
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 3.6,
|
| 259 |
-
"learning_rate": 9.
|
| 260 |
-
"loss": 0.
|
| 261 |
-
"step":
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 3.7,
|
| 265 |
-
"learning_rate": 9.
|
| 266 |
-
"loss": 0.
|
| 267 |
-
"step":
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 3.8,
|
| 271 |
-
"learning_rate": 8.
|
| 272 |
-
"loss": 0.
|
| 273 |
-
"step":
|
| 274 |
},
|
| 275 |
{
|
| 276 |
"epoch": 3.9,
|
| 277 |
-
"learning_rate": 8.
|
| 278 |
-
"loss": 0.
|
| 279 |
-
"step":
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 4.0,
|
| 283 |
-
"eval_loss": 0.
|
| 284 |
"eval_max_distance": 4,
|
| 285 |
"eval_mean_distance": 0,
|
| 286 |
-
"eval_runtime":
|
| 287 |
-
"eval_samples_per_second":
|
| 288 |
-
"eval_steps_per_second":
|
| 289 |
-
"step":
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 4.0,
|
| 293 |
-
"learning_rate": 8.
|
| 294 |
-
"loss": 0.
|
| 295 |
-
"step":
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 4.1,
|
| 299 |
-
"learning_rate": 8.
|
| 300 |
-
"loss": 0.
|
| 301 |
-
"step":
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 4.2,
|
| 305 |
-
"learning_rate": 8.
|
| 306 |
-
"loss": 0.
|
| 307 |
-
"step":
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 4.3,
|
| 311 |
-
"learning_rate": 8.
|
| 312 |
-
"loss": 0.
|
| 313 |
-
"step":
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 4.4,
|
| 317 |
-
"learning_rate": 8.
|
| 318 |
-
"loss": 0.
|
| 319 |
-
"step":
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 4.5,
|
| 323 |
-
"learning_rate": 8.
|
| 324 |
-
"loss": 0.
|
| 325 |
-
"step":
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 4.6,
|
| 329 |
-
"learning_rate": 8.
|
| 330 |
-
"loss": 0.
|
| 331 |
-
"step":
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 4.7,
|
| 335 |
-
"learning_rate": 8.
|
| 336 |
-
"loss": 0.
|
| 337 |
-
"step":
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 4.8,
|
| 341 |
-
"learning_rate": 8.
|
| 342 |
-
"loss": 0.
|
| 343 |
-
"step":
|
| 344 |
},
|
| 345 |
{
|
| 346 |
"epoch": 4.9,
|
| 347 |
-
"learning_rate": 8.
|
| 348 |
-
"loss": 0.
|
| 349 |
-
"step":
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 5.0,
|
| 353 |
-
"eval_loss": 0.
|
| 354 |
"eval_max_distance": 4,
|
| 355 |
"eval_mean_distance": 0,
|
| 356 |
-
"eval_runtime":
|
| 357 |
-
"eval_samples_per_second":
|
| 358 |
-
"eval_steps_per_second": 1.
|
| 359 |
-
"step":
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 5.0,
|
| 363 |
-
"learning_rate": 8.
|
| 364 |
-
"loss": 0.
|
| 365 |
-
"step":
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 5.1,
|
| 369 |
-
"learning_rate": 8.
|
| 370 |
-
"loss": 0.
|
| 371 |
-
"step":
|
| 372 |
},
|
| 373 |
{
|
| 374 |
"epoch": 5.2,
|
| 375 |
-
"learning_rate": 8.
|
| 376 |
-
"loss": 0.
|
| 377 |
-
"step":
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 5.3,
|
| 381 |
-
"learning_rate": 8.
|
| 382 |
-
"loss": 0.
|
| 383 |
-
"step":
|
| 384 |
},
|
| 385 |
{
|
| 386 |
"epoch": 5.4,
|
| 387 |
-
"learning_rate": 8.
|
| 388 |
-
"loss": 0.
|
| 389 |
-
"step":
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 5.5,
|
| 393 |
-
"learning_rate": 8.
|
| 394 |
-
"loss": 0.
|
| 395 |
-
"step":
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 5.6,
|
| 399 |
-
"learning_rate": 7.
|
| 400 |
-
"loss": 0.
|
| 401 |
-
"step":
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 5.7,
|
| 405 |
-
"learning_rate": 7.
|
| 406 |
-
"loss": 0.
|
| 407 |
-
"step":
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 5.8,
|
| 411 |
-
"learning_rate": 7.
|
| 412 |
-
"loss": 0.
|
| 413 |
-
"step":
|
| 414 |
},
|
| 415 |
{
|
| 416 |
"epoch": 5.9,
|
| 417 |
-
"learning_rate": 7.
|
| 418 |
-
"loss": 0.
|
| 419 |
-
"step":
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 6.0,
|
| 423 |
-
"eval_loss": 0.
|
| 424 |
-
"eval_max_distance":
|
| 425 |
"eval_mean_distance": 0,
|
| 426 |
-
"eval_runtime":
|
| 427 |
-
"eval_samples_per_second":
|
| 428 |
-
"eval_steps_per_second": 1.
|
| 429 |
-
"step":
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 6.0,
|
| 433 |
-
"learning_rate": 7.
|
| 434 |
-
"loss": 0.
|
| 435 |
-
"step":
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 6.1,
|
| 439 |
-
"learning_rate": 7.
|
| 440 |
-
"loss": 0.
|
| 441 |
-
"step":
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 6.2,
|
| 445 |
-
"learning_rate": 7.
|
| 446 |
-
"loss": 0.
|
| 447 |
-
"step":
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 6.3,
|
| 451 |
-
"learning_rate": 7.
|
| 452 |
-
"loss": 0.
|
| 453 |
-
"step":
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 6.4,
|
| 457 |
-
"learning_rate": 7.
|
| 458 |
-
"loss": 0.
|
| 459 |
-
"step":
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 6.5,
|
| 463 |
-
"learning_rate": 7.
|
| 464 |
-
"loss": 0.
|
| 465 |
-
"step":
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 6.6,
|
| 469 |
-
"learning_rate": 7.
|
| 470 |
-
"loss": 0.
|
| 471 |
-
"step":
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 6.7,
|
| 475 |
-
"learning_rate": 7.
|
| 476 |
-
"loss": 0.
|
| 477 |
-
"step":
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 6.8,
|
| 481 |
-
"learning_rate": 7.
|
| 482 |
-
"loss": 0.
|
| 483 |
-
"step":
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 6.9,
|
| 487 |
-
"learning_rate": 7.
|
| 488 |
-
"loss": 0.
|
| 489 |
-
"step":
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 7.0,
|
| 493 |
-
"eval_loss": 0.
|
| 494 |
-
"eval_max_distance":
|
| 495 |
"eval_mean_distance": 0,
|
| 496 |
-
"eval_runtime":
|
| 497 |
-
"eval_samples_per_second":
|
| 498 |
-
"eval_steps_per_second": 1.
|
| 499 |
-
"step":
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 7.0,
|
| 503 |
-
"learning_rate": 7.
|
| 504 |
-
"loss": 0.
|
| 505 |
-
"step":
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 7.1,
|
| 509 |
-
"learning_rate": 7.
|
| 510 |
-
"loss": 0.
|
| 511 |
-
"step":
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 7.2,
|
| 515 |
-
"learning_rate": 7.
|
| 516 |
-
"loss": 0.
|
| 517 |
-
"step":
|
| 518 |
},
|
| 519 |
{
|
| 520 |
"epoch": 7.3,
|
| 521 |
-
"learning_rate": 7.
|
| 522 |
-
"loss": 0.
|
| 523 |
-
"step":
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 7.4,
|
| 527 |
-
"learning_rate": 6.
|
| 528 |
-
"loss": 0.
|
| 529 |
-
"step":
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 7.5,
|
| 533 |
-
"learning_rate": 6.
|
| 534 |
-
"loss": 0.
|
| 535 |
-
"step":
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"epoch": 7.6,
|
| 539 |
-
"learning_rate": 6.
|
| 540 |
-
"loss": 0.
|
| 541 |
-
"step":
|
| 542 |
},
|
| 543 |
{
|
| 544 |
"epoch": 7.7,
|
| 545 |
-
"learning_rate": 6.
|
| 546 |
-
"loss": 0.
|
| 547 |
-
"step":
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 7.8,
|
| 551 |
-
"learning_rate": 6.
|
| 552 |
-
"loss": 0.
|
| 553 |
-
"step":
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 7.9,
|
| 557 |
-
"learning_rate": 6.
|
| 558 |
-
"loss": 0.
|
| 559 |
-
"step":
|
| 560 |
},
|
| 561 |
{
|
| 562 |
"epoch": 8.0,
|
| 563 |
-
"eval_loss": 0.
|
| 564 |
-
"eval_max_distance":
|
| 565 |
"eval_mean_distance": 0,
|
| 566 |
-
"eval_runtime":
|
| 567 |
-
"eval_samples_per_second":
|
| 568 |
-
"eval_steps_per_second": 1.
|
| 569 |
-
"step":
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 8.0,
|
| 573 |
-
"learning_rate": 6.
|
| 574 |
-
"loss": 0.
|
| 575 |
-
"step":
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 8.1,
|
| 579 |
-
"learning_rate": 6.
|
| 580 |
-
"loss": 0.
|
| 581 |
-
"step":
|
| 582 |
},
|
| 583 |
{
|
| 584 |
"epoch": 8.2,
|
| 585 |
-
"learning_rate": 6.
|
| 586 |
-
"loss": 0.
|
| 587 |
-
"step":
|
| 588 |
},
|
| 589 |
{
|
| 590 |
"epoch": 8.3,
|
| 591 |
-
"learning_rate": 6.
|
| 592 |
-
"loss": 0.
|
| 593 |
-
"step":
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 8.4,
|
| 597 |
-
"learning_rate": 6.
|
| 598 |
-
"loss": 0.
|
| 599 |
-
"step":
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 8.5,
|
| 603 |
-
"learning_rate": 6.
|
| 604 |
-
"loss": 0.
|
| 605 |
-
"step":
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 8.6,
|
| 609 |
-
"learning_rate": 6.
|
| 610 |
-
"loss": 0.
|
| 611 |
-
"step":
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 8.7,
|
| 615 |
-
"learning_rate": 6.
|
| 616 |
-
"loss": 0.
|
| 617 |
-
"step":
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 8.8,
|
| 621 |
-
"learning_rate": 6.
|
| 622 |
-
"loss": 0.
|
| 623 |
-
"step":
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 8.9,
|
| 627 |
-
"learning_rate": 6.
|
| 628 |
-
"loss": 0.
|
| 629 |
-
"step":
|
| 630 |
},
|
| 631 |
{
|
| 632 |
"epoch": 9.0,
|
| 633 |
-
"eval_loss": 0.
|
| 634 |
"eval_max_distance": 4,
|
| 635 |
"eval_mean_distance": 0,
|
| 636 |
-
"eval_runtime":
|
| 637 |
-
"eval_samples_per_second":
|
| 638 |
-
"eval_steps_per_second": 1.
|
| 639 |
-
"step":
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 9.0,
|
| 643 |
-
"learning_rate": 6.
|
| 644 |
-
"loss": 0.
|
| 645 |
-
"step":
|
| 646 |
},
|
| 647 |
{
|
| 648 |
"epoch": 9.1,
|
| 649 |
-
"learning_rate": 6.
|
| 650 |
-
"loss": 0.
|
| 651 |
-
"step":
|
| 652 |
},
|
| 653 |
{
|
| 654 |
"epoch": 9.2,
|
| 655 |
-
"learning_rate": 5.
|
| 656 |
-
"loss": 0.
|
| 657 |
-
"step":
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 9.3,
|
| 661 |
-
"learning_rate": 5.
|
| 662 |
-
"loss": 0.
|
| 663 |
-
"step":
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 9.4,
|
| 667 |
-
"learning_rate": 5.
|
| 668 |
-
"loss": 0.
|
| 669 |
-
"step":
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 9.5,
|
| 673 |
-
"learning_rate": 5.
|
| 674 |
-
"loss": 0.
|
| 675 |
-
"step":
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 9.6,
|
| 679 |
-
"learning_rate": 5.
|
| 680 |
-
"loss": 0.
|
| 681 |
-
"step":
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 9.7,
|
| 685 |
-
"learning_rate": 5.
|
| 686 |
-
"loss": 0.
|
| 687 |
-
"step":
|
| 688 |
},
|
| 689 |
{
|
| 690 |
"epoch": 9.8,
|
| 691 |
-
"learning_rate": 5.
|
| 692 |
-
"loss": 0.
|
| 693 |
-
"step":
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 9.9,
|
| 697 |
-
"learning_rate": 5.
|
| 698 |
-
"loss": 0.
|
| 699 |
-
"step":
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 10.0,
|
| 703 |
-
"eval_loss": 0.
|
| 704 |
-
"eval_max_distance":
|
| 705 |
"eval_mean_distance": 0,
|
| 706 |
-
"eval_runtime":
|
| 707 |
-
"eval_samples_per_second":
|
| 708 |
-
"eval_steps_per_second": 1.
|
| 709 |
-
"step":
|
| 710 |
},
|
| 711 |
{
|
| 712 |
"epoch": 10.0,
|
| 713 |
-
"learning_rate": 5.
|
| 714 |
-
"loss": 0.
|
| 715 |
-
"step":
|
| 716 |
},
|
| 717 |
{
|
| 718 |
"epoch": 10.1,
|
| 719 |
-
"learning_rate": 5.
|
| 720 |
-
"loss": 0.
|
| 721 |
-
"step":
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 10.2,
|
| 725 |
-
"learning_rate": 5.
|
| 726 |
-
"loss": 0.
|
| 727 |
-
"step":
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 10.3,
|
| 731 |
-
"learning_rate": 5.
|
| 732 |
-
"loss": 0.
|
| 733 |
-
"step":
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 10.4,
|
| 737 |
-
"learning_rate": 5.
|
| 738 |
-
"loss": 0.
|
| 739 |
-
"step":
|
| 740 |
},
|
| 741 |
{
|
| 742 |
"epoch": 10.5,
|
| 743 |
-
"learning_rate": 5.
|
| 744 |
-
"loss": 0.
|
| 745 |
-
"step":
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 10.6,
|
| 749 |
-
"learning_rate": 5.
|
| 750 |
-
"loss": 0.
|
| 751 |
-
"step":
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 10.7,
|
| 755 |
-
"learning_rate": 5.
|
| 756 |
-
"loss": 0.
|
| 757 |
-
"step":
|
| 758 |
},
|
| 759 |
{
|
| 760 |
"epoch": 10.8,
|
| 761 |
-
"learning_rate": 5.
|
| 762 |
-
"loss": 0.
|
| 763 |
-
"step":
|
| 764 |
},
|
| 765 |
{
|
| 766 |
"epoch": 10.9,
|
| 767 |
-
"learning_rate": 5.
|
| 768 |
-
"loss": 0.
|
| 769 |
-
"step":
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 11.0,
|
| 773 |
-
"eval_loss": 0.
|
| 774 |
-
"eval_max_distance":
|
| 775 |
"eval_mean_distance": 0,
|
| 776 |
-
"eval_runtime":
|
| 777 |
-
"eval_samples_per_second":
|
| 778 |
-
"eval_steps_per_second": 1.
|
| 779 |
-
"step":
|
| 780 |
},
|
| 781 |
{
|
| 782 |
"epoch": 11.0,
|
| 783 |
-
"learning_rate": 4.
|
| 784 |
-
"loss": 0.
|
| 785 |
-
"step":
|
| 786 |
},
|
| 787 |
{
|
| 788 |
"epoch": 11.1,
|
| 789 |
-
"learning_rate": 4.
|
| 790 |
-
"loss": 0.
|
| 791 |
-
"step":
|
| 792 |
},
|
| 793 |
{
|
| 794 |
"epoch": 11.2,
|
| 795 |
-
"learning_rate": 4.
|
| 796 |
-
"loss": 0.
|
| 797 |
-
"step":
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 11.3,
|
| 801 |
-
"learning_rate": 4.
|
| 802 |
-
"loss": 0.
|
| 803 |
-
"step":
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 11.4,
|
| 807 |
-
"learning_rate": 4.
|
| 808 |
-
"loss": 0.
|
| 809 |
-
"step":
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 11.5,
|
| 813 |
-
"learning_rate": 4.
|
| 814 |
-
"loss": 0.
|
| 815 |
-
"step":
|
| 816 |
},
|
| 817 |
{
|
| 818 |
"epoch": 11.6,
|
| 819 |
-
"learning_rate": 4.
|
| 820 |
-
"loss": 0.
|
| 821 |
-
"step":
|
| 822 |
},
|
| 823 |
{
|
| 824 |
"epoch": 11.7,
|
| 825 |
-
"learning_rate": 4.
|
| 826 |
-
"loss": 0.
|
| 827 |
-
"step":
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 11.8,
|
| 831 |
-
"learning_rate": 4.
|
| 832 |
-
"loss": 0.
|
| 833 |
-
"step":
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 11.9,
|
| 837 |
-
"learning_rate": 4.
|
| 838 |
-
"loss": 0.
|
| 839 |
-
"step":
|
| 840 |
},
|
| 841 |
{
|
| 842 |
"epoch": 12.0,
|
| 843 |
-
"eval_loss": 0.
|
| 844 |
"eval_max_distance": 3,
|
| 845 |
"eval_mean_distance": 0,
|
| 846 |
-
"eval_runtime":
|
| 847 |
-
"eval_samples_per_second":
|
| 848 |
-
"eval_steps_per_second": 1.
|
| 849 |
-
"step":
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 12.0,
|
| 853 |
-
"learning_rate": 4.
|
| 854 |
-
"loss": 0.
|
| 855 |
-
"step":
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 12.1,
|
| 859 |
-
"learning_rate": 4.
|
| 860 |
-
"loss": 0.
|
| 861 |
-
"step":
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 12.2,
|
| 865 |
-
"learning_rate": 4.
|
| 866 |
-
"loss": 0.
|
| 867 |
-
"step":
|
| 868 |
},
|
| 869 |
{
|
| 870 |
"epoch": 12.3,
|
| 871 |
-
"learning_rate": 4.
|
| 872 |
-
"loss": 0.
|
| 873 |
-
"step":
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 12.4,
|
| 877 |
-
"learning_rate": 4.
|
| 878 |
-
"loss": 0.
|
| 879 |
-
"step":
|
| 880 |
},
|
| 881 |
{
|
| 882 |
"epoch": 12.5,
|
| 883 |
-
"learning_rate": 4.
|
| 884 |
-
"loss": 0.
|
| 885 |
-
"step":
|
| 886 |
},
|
| 887 |
{
|
| 888 |
"epoch": 12.6,
|
| 889 |
-
"learning_rate": 4.
|
| 890 |
-
"loss": 0.
|
| 891 |
-
"step":
|
| 892 |
},
|
| 893 |
{
|
| 894 |
"epoch": 12.7,
|
| 895 |
-
"learning_rate": 4.
|
| 896 |
-
"loss": 0.
|
| 897 |
-
"step":
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 12.8,
|
| 901 |
-
"learning_rate": 3.
|
| 902 |
-
"loss": 0.
|
| 903 |
-
"step":
|
| 904 |
},
|
| 905 |
{
|
| 906 |
"epoch": 12.9,
|
| 907 |
-
"learning_rate": 3.
|
| 908 |
-
"loss": 0.
|
| 909 |
-
"step":
|
| 910 |
},
|
| 911 |
{
|
| 912 |
"epoch": 13.0,
|
| 913 |
-
"eval_loss": 0.
|
| 914 |
-
"eval_max_distance":
|
| 915 |
"eval_mean_distance": 0,
|
| 916 |
-
"eval_runtime":
|
| 917 |
-
"eval_samples_per_second":
|
| 918 |
-
"eval_steps_per_second": 1.
|
| 919 |
-
"step":
|
| 920 |
},
|
| 921 |
{
|
| 922 |
"epoch": 13.0,
|
| 923 |
-
"learning_rate": 3.
|
| 924 |
-
"loss": 0.
|
| 925 |
-
"step":
|
| 926 |
},
|
| 927 |
{
|
| 928 |
"epoch": 13.1,
|
| 929 |
-
"learning_rate": 3.
|
| 930 |
-
"loss": 0.
|
| 931 |
-
"step":
|
| 932 |
},
|
| 933 |
{
|
| 934 |
"epoch": 13.2,
|
| 935 |
-
"learning_rate": 3.
|
| 936 |
-
"loss": 0.
|
| 937 |
-
"step":
|
| 938 |
},
|
| 939 |
{
|
| 940 |
"epoch": 13.3,
|
| 941 |
-
"learning_rate": 3.
|
| 942 |
-
"loss": 0.
|
| 943 |
-
"step":
|
| 944 |
},
|
| 945 |
{
|
| 946 |
"epoch": 13.4,
|
| 947 |
-
"learning_rate": 3.
|
| 948 |
-
"loss": 0.
|
| 949 |
-
"step":
|
| 950 |
},
|
| 951 |
{
|
| 952 |
"epoch": 13.5,
|
| 953 |
-
"learning_rate": 3.
|
| 954 |
-
"loss": 0.
|
| 955 |
-
"step":
|
| 956 |
},
|
| 957 |
{
|
| 958 |
"epoch": 13.6,
|
| 959 |
-
"learning_rate": 3.
|
| 960 |
-
"loss": 0.
|
| 961 |
-
"step":
|
| 962 |
},
|
| 963 |
{
|
| 964 |
"epoch": 13.7,
|
| 965 |
-
"learning_rate":
|
| 966 |
-
"loss": 0.
|
| 967 |
-
"step":
|
| 968 |
},
|
| 969 |
{
|
| 970 |
"epoch": 13.8,
|
| 971 |
-
"learning_rate":
|
| 972 |
-
"loss": 0.
|
| 973 |
-
"step":
|
| 974 |
},
|
| 975 |
{
|
| 976 |
"epoch": 13.9,
|
| 977 |
-
"learning_rate":
|
| 978 |
-
"loss": 0.
|
| 979 |
-
"step":
|
| 980 |
},
|
| 981 |
{
|
| 982 |
"epoch": 14.0,
|
| 983 |
-
"eval_loss": 0.
|
| 984 |
-
"eval_max_distance":
|
| 985 |
"eval_mean_distance": 0,
|
| 986 |
-
"eval_runtime":
|
| 987 |
-
"eval_samples_per_second":
|
| 988 |
-
"eval_steps_per_second":
|
| 989 |
-
"step":
|
| 990 |
},
|
| 991 |
{
|
| 992 |
"epoch": 14.0,
|
| 993 |
-
"
|
| 994 |
-
"
|
| 995 |
-
"
|
| 996 |
-
|
| 997 |
-
|
| 998 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 999 |
}
|
| 1000 |
],
|
| 1001 |
-
"logging_steps":
|
| 1002 |
-
"max_steps":
|
| 1003 |
-
"num_train_epochs":
|
| 1004 |
-
"save_steps":
|
| 1005 |
-
"total_flos":
|
| 1006 |
"trial_name": null,
|
| 1007 |
"trial_params": null
|
| 1008 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 462780,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.0,
|
| 13 |
+
"learning_rate": 2.160853969488742e-09,
|
| 14 |
+
"loss": 0.0,
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.1,
|
| 19 |
+
"learning_rate": 5.000216085396949e-06,
|
| 20 |
+
"loss": 0.0011,
|
| 21 |
+
"step": 2314
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 1.0000432170793898e-05,
|
| 26 |
+
"loss": 0.001,
|
| 27 |
+
"step": 4628
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.3,
|
| 31 |
+
"learning_rate": 1.5000648256190847e-05,
|
| 32 |
+
"loss": 0.0011,
|
| 33 |
+
"step": 6942
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.4,
|
| 37 |
+
"learning_rate": 2.0000864341587796e-05,
|
| 38 |
+
"loss": 0.0011,
|
| 39 |
+
"step": 9256
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.5,
|
| 43 |
+
"learning_rate": 2.5001080426984747e-05,
|
| 44 |
+
"loss": 0.001,
|
| 45 |
+
"step": 11570
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.6,
|
| 49 |
+
"learning_rate": 3.0001296512381694e-05,
|
| 50 |
+
"loss": 0.0011,
|
| 51 |
+
"step": 13884
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.7,
|
| 55 |
+
"learning_rate": 3.500151259777865e-05,
|
| 56 |
+
"loss": 0.001,
|
| 57 |
+
"step": 16198
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.8,
|
| 61 |
+
"learning_rate": 4.000172868317559e-05,
|
| 62 |
+
"loss": 0.0011,
|
| 63 |
+
"step": 18512
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.9,
|
| 67 |
+
"learning_rate": 4.5001944768572543e-05,
|
| 68 |
+
"loss": 0.0011,
|
| 69 |
+
"step": 20826
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 1.0,
|
| 73 |
+
"eval_loss": 0.006049504969269037,
|
| 74 |
+
"eval_max_distance": 4,
|
| 75 |
"eval_mean_distance": 0,
|
| 76 |
+
"eval_runtime": 12.5214,
|
| 77 |
+
"eval_samples_per_second": 20.605,
|
| 78 |
+
"eval_steps_per_second": 1.438,
|
| 79 |
+
"step": 23139
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.0,
|
| 83 |
+
"learning_rate": 5.0002160853969494e-05,
|
| 84 |
+
"loss": 0.001,
|
| 85 |
+
"step": 23140
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"epoch": 1.1,
|
| 89 |
+
"learning_rate": 5.500237693936644e-05,
|
| 90 |
+
"loss": 0.0011,
|
| 91 |
+
"step": 25454
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 1.2,
|
| 95 |
+
"learning_rate": 6.000259302476339e-05,
|
| 96 |
+
"loss": 0.001,
|
| 97 |
+
"step": 27768
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 1.3,
|
| 101 |
+
"learning_rate": 6.500280911016034e-05,
|
| 102 |
+
"loss": 0.0011,
|
| 103 |
+
"step": 30082
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 1.4,
|
| 107 |
+
"learning_rate": 7.00030251955573e-05,
|
| 108 |
+
"loss": 0.0011,
|
| 109 |
+
"step": 32396
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 1.5,
|
| 113 |
+
"learning_rate": 7.500324128095424e-05,
|
| 114 |
+
"loss": 0.0012,
|
| 115 |
+
"step": 34710
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 1.6,
|
| 119 |
+
"learning_rate": 8.000345736635119e-05,
|
| 120 |
+
"loss": 0.0012,
|
| 121 |
+
"step": 37024
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 1.7,
|
| 125 |
+
"learning_rate": 8.500367345174813e-05,
|
| 126 |
+
"loss": 0.0013,
|
| 127 |
+
"step": 39338
|
| 128 |
},
|
| 129 |
{
|
| 130 |
"epoch": 1.8,
|
| 131 |
+
"learning_rate": 9.000388953714509e-05,
|
| 132 |
+
"loss": 0.0013,
|
| 133 |
+
"step": 41652
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 1.9,
|
| 137 |
+
"learning_rate": 9.500410562254203e-05,
|
| 138 |
+
"loss": 0.0013,
|
| 139 |
+
"step": 43966
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 2.0,
|
| 143 |
+
"eval_loss": 0.0038143955171108246,
|
| 144 |
+
"eval_max_distance": 4,
|
| 145 |
"eval_mean_distance": 0,
|
| 146 |
+
"eval_runtime": 12.0107,
|
| 147 |
+
"eval_samples_per_second": 21.481,
|
| 148 |
+
"eval_steps_per_second": 1.499,
|
| 149 |
+
"step": 46278
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 2.0,
|
| 153 |
+
"learning_rate": 9.9999519810229e-05,
|
| 154 |
+
"loss": 0.0013,
|
| 155 |
+
"step": 46280
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 2.1,
|
| 159 |
+
"learning_rate": 9.94439402451849e-05,
|
| 160 |
+
"loss": 0.0013,
|
| 161 |
+
"step": 48594
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 2.2,
|
| 165 |
+
"learning_rate": 9.88883606801408e-05,
|
| 166 |
+
"loss": 0.0013,
|
| 167 |
+
"step": 50908
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 2.3,
|
| 171 |
+
"learning_rate": 9.83327811150967e-05,
|
| 172 |
+
"loss": 0.0013,
|
| 173 |
+
"step": 53222
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 2.4,
|
| 177 |
+
"learning_rate": 9.777720155005259e-05,
|
| 178 |
+
"loss": 0.0014,
|
| 179 |
+
"step": 55536
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 2.5,
|
| 183 |
+
"learning_rate": 9.722162198500848e-05,
|
| 184 |
+
"loss": 0.0014,
|
| 185 |
+
"step": 57850
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 2.6,
|
| 189 |
+
"learning_rate": 9.666604241996438e-05,
|
| 190 |
+
"loss": 0.0014,
|
| 191 |
+
"step": 60164
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 2.7,
|
| 195 |
+
"learning_rate": 9.611046285492027e-05,
|
| 196 |
+
"loss": 0.0013,
|
| 197 |
+
"step": 62478
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 2.8,
|
| 201 |
+
"learning_rate": 9.555488328987616e-05,
|
| 202 |
+
"loss": 0.0014,
|
| 203 |
+
"step": 64792
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 2.9,
|
| 207 |
+
"learning_rate": 9.499930372483206e-05,
|
| 208 |
+
"loss": 0.0014,
|
| 209 |
+
"step": 67106
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 3.0,
|
| 213 |
+
"eval_loss": 0.003588364226743579,
|
| 214 |
+
"eval_max_distance": 4,
|
| 215 |
"eval_mean_distance": 0,
|
| 216 |
+
"eval_runtime": 11.9267,
|
| 217 |
+
"eval_samples_per_second": 21.632,
|
| 218 |
+
"eval_steps_per_second": 1.509,
|
| 219 |
+
"step": 69417
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 3.0,
|
| 223 |
+
"learning_rate": 9.444372415978795e-05,
|
| 224 |
+
"loss": 0.0015,
|
| 225 |
+
"step": 69420
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 3.1,
|
| 229 |
+
"learning_rate": 9.388814459474385e-05,
|
| 230 |
+
"loss": 0.0013,
|
| 231 |
+
"step": 71734
|
| 232 |
},
|
| 233 |
{
|
| 234 |
"epoch": 3.2,
|
| 235 |
+
"learning_rate": 9.333256502969974e-05,
|
| 236 |
+
"loss": 0.0014,
|
| 237 |
+
"step": 74048
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 3.3,
|
| 241 |
+
"learning_rate": 9.277698546465565e-05,
|
| 242 |
+
"loss": 0.0014,
|
| 243 |
+
"step": 76362
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 3.4,
|
| 247 |
+
"learning_rate": 9.222140589961154e-05,
|
| 248 |
+
"loss": 0.0014,
|
| 249 |
+
"step": 78676
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 3.5,
|
| 253 |
+
"learning_rate": 9.166582633456743e-05,
|
| 254 |
+
"loss": 0.0014,
|
| 255 |
+
"step": 80990
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 3.6,
|
| 259 |
+
"learning_rate": 9.111024676952333e-05,
|
| 260 |
+
"loss": 0.0014,
|
| 261 |
+
"step": 83304
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 3.7,
|
| 265 |
+
"learning_rate": 9.055466720447922e-05,
|
| 266 |
+
"loss": 0.0014,
|
| 267 |
+
"step": 85618
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 3.8,
|
| 271 |
+
"learning_rate": 8.999908763943511e-05,
|
| 272 |
+
"loss": 0.0014,
|
| 273 |
+
"step": 87932
|
| 274 |
},
|
| 275 |
{
|
| 276 |
"epoch": 3.9,
|
| 277 |
+
"learning_rate": 8.944350807439101e-05,
|
| 278 |
+
"loss": 0.0014,
|
| 279 |
+
"step": 90246
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 4.0,
|
| 283 |
+
"eval_loss": 0.00436835503205657,
|
| 284 |
"eval_max_distance": 4,
|
| 285 |
"eval_mean_distance": 0,
|
| 286 |
+
"eval_runtime": 12.0127,
|
| 287 |
+
"eval_samples_per_second": 21.477,
|
| 288 |
+
"eval_steps_per_second": 1.498,
|
| 289 |
+
"step": 92556
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 4.0,
|
| 293 |
+
"learning_rate": 8.88879285093469e-05,
|
| 294 |
+
"loss": 0.0014,
|
| 295 |
+
"step": 92560
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 4.1,
|
| 299 |
+
"learning_rate": 8.83323489443028e-05,
|
| 300 |
+
"loss": 0.0013,
|
| 301 |
+
"step": 94874
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 4.2,
|
| 305 |
+
"learning_rate": 8.777676937925869e-05,
|
| 306 |
+
"loss": 0.0013,
|
| 307 |
+
"step": 97188
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 4.3,
|
| 311 |
+
"learning_rate": 8.722118981421458e-05,
|
| 312 |
+
"loss": 0.0014,
|
| 313 |
+
"step": 99502
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 4.4,
|
| 317 |
+
"learning_rate": 8.666561024917048e-05,
|
| 318 |
+
"loss": 0.0012,
|
| 319 |
+
"step": 101816
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 4.5,
|
| 323 |
+
"learning_rate": 8.611003068412637e-05,
|
| 324 |
+
"loss": 0.0014,
|
| 325 |
+
"step": 104130
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 4.6,
|
| 329 |
+
"learning_rate": 8.555445111908226e-05,
|
| 330 |
+
"loss": 0.0013,
|
| 331 |
+
"step": 106444
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 4.7,
|
| 335 |
+
"learning_rate": 8.499887155403816e-05,
|
| 336 |
+
"loss": 0.0013,
|
| 337 |
+
"step": 108758
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 4.8,
|
| 341 |
+
"learning_rate": 8.444329198899405e-05,
|
| 342 |
+
"loss": 0.0014,
|
| 343 |
+
"step": 111072
|
| 344 |
},
|
| 345 |
{
|
| 346 |
"epoch": 4.9,
|
| 347 |
+
"learning_rate": 8.388771242394994e-05,
|
| 348 |
+
"loss": 0.0013,
|
| 349 |
+
"step": 113386
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 5.0,
|
| 353 |
+
"eval_loss": 0.005551120266318321,
|
| 354 |
"eval_max_distance": 4,
|
| 355 |
"eval_mean_distance": 0,
|
| 356 |
+
"eval_runtime": 11.9604,
|
| 357 |
+
"eval_samples_per_second": 21.571,
|
| 358 |
+
"eval_steps_per_second": 1.505,
|
| 359 |
+
"step": 115695
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 5.0,
|
| 363 |
+
"learning_rate": 8.333213285890584e-05,
|
| 364 |
+
"loss": 0.0014,
|
| 365 |
+
"step": 115700
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 5.1,
|
| 369 |
+
"learning_rate": 8.277655329386173e-05,
|
| 370 |
+
"loss": 0.0012,
|
| 371 |
+
"step": 118014
|
| 372 |
},
|
| 373 |
{
|
| 374 |
"epoch": 5.2,
|
| 375 |
+
"learning_rate": 8.222097372881762e-05,
|
| 376 |
+
"loss": 0.0013,
|
| 377 |
+
"step": 120328
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 5.3,
|
| 381 |
+
"learning_rate": 8.166539416377352e-05,
|
| 382 |
+
"loss": 0.0013,
|
| 383 |
+
"step": 122642
|
| 384 |
},
|
| 385 |
{
|
| 386 |
"epoch": 5.4,
|
| 387 |
+
"learning_rate": 8.110981459872941e-05,
|
| 388 |
+
"loss": 0.0013,
|
| 389 |
+
"step": 124956
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 5.5,
|
| 393 |
+
"learning_rate": 8.055423503368532e-05,
|
| 394 |
+
"loss": 0.0012,
|
| 395 |
+
"step": 127270
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 5.6,
|
| 399 |
+
"learning_rate": 7.999865546864121e-05,
|
| 400 |
+
"loss": 0.0013,
|
| 401 |
+
"step": 129584
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 5.7,
|
| 405 |
+
"learning_rate": 7.94430759035971e-05,
|
| 406 |
+
"loss": 0.0014,
|
| 407 |
+
"step": 131898
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 5.8,
|
| 411 |
+
"learning_rate": 7.8887496338553e-05,
|
| 412 |
+
"loss": 0.0013,
|
| 413 |
+
"step": 134212
|
| 414 |
},
|
| 415 |
{
|
| 416 |
"epoch": 5.9,
|
| 417 |
+
"learning_rate": 7.83319167735089e-05,
|
| 418 |
+
"loss": 0.0014,
|
| 419 |
+
"step": 136526
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 6.0,
|
| 423 |
+
"eval_loss": 0.0034297292586416006,
|
| 424 |
+
"eval_max_distance": 2,
|
| 425 |
"eval_mean_distance": 0,
|
| 426 |
+
"eval_runtime": 11.9347,
|
| 427 |
+
"eval_samples_per_second": 21.618,
|
| 428 |
+
"eval_steps_per_second": 1.508,
|
| 429 |
+
"step": 138834
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 6.0,
|
| 433 |
+
"learning_rate": 7.777633720846479e-05,
|
| 434 |
+
"loss": 0.0014,
|
| 435 |
+
"step": 138840
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 6.1,
|
| 439 |
+
"learning_rate": 7.722075764342068e-05,
|
| 440 |
+
"loss": 0.0012,
|
| 441 |
+
"step": 141154
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 6.2,
|
| 445 |
+
"learning_rate": 7.666517807837657e-05,
|
| 446 |
+
"loss": 0.0012,
|
| 447 |
+
"step": 143468
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 6.3,
|
| 451 |
+
"learning_rate": 7.610959851333247e-05,
|
| 452 |
+
"loss": 0.0012,
|
| 453 |
+
"step": 145782
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 6.4,
|
| 457 |
+
"learning_rate": 7.555401894828836e-05,
|
| 458 |
+
"loss": 0.0012,
|
| 459 |
+
"step": 148096
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 6.5,
|
| 463 |
+
"learning_rate": 7.499843938324425e-05,
|
| 464 |
+
"loss": 0.0013,
|
| 465 |
+
"step": 150410
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 6.6,
|
| 469 |
+
"learning_rate": 7.444285981820016e-05,
|
| 470 |
+
"loss": 0.0012,
|
| 471 |
+
"step": 152724
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 6.7,
|
| 475 |
+
"learning_rate": 7.388728025315606e-05,
|
| 476 |
+
"loss": 0.0012,
|
| 477 |
+
"step": 155038
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 6.8,
|
| 481 |
+
"learning_rate": 7.333170068811195e-05,
|
| 482 |
+
"loss": 0.0013,
|
| 483 |
+
"step": 157352
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 6.9,
|
| 487 |
+
"learning_rate": 7.277612112306784e-05,
|
| 488 |
+
"loss": 0.0012,
|
| 489 |
+
"step": 159666
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 7.0,
|
| 493 |
+
"eval_loss": 0.0036055969540029764,
|
| 494 |
+
"eval_max_distance": 2,
|
| 495 |
"eval_mean_distance": 0,
|
| 496 |
+
"eval_runtime": 11.9356,
|
| 497 |
+
"eval_samples_per_second": 21.616,
|
| 498 |
+
"eval_steps_per_second": 1.508,
|
| 499 |
+
"step": 161973
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 7.0,
|
| 503 |
+
"learning_rate": 7.222054155802374e-05,
|
| 504 |
+
"loss": 0.0012,
|
| 505 |
+
"step": 161980
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 7.1,
|
| 509 |
+
"learning_rate": 7.166496199297963e-05,
|
| 510 |
+
"loss": 0.0011,
|
| 511 |
+
"step": 164294
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 7.2,
|
| 515 |
+
"learning_rate": 7.110938242793552e-05,
|
| 516 |
+
"loss": 0.0011,
|
| 517 |
+
"step": 166608
|
| 518 |
},
|
| 519 |
{
|
| 520 |
"epoch": 7.3,
|
| 521 |
+
"learning_rate": 7.055380286289142e-05,
|
| 522 |
+
"loss": 0.0013,
|
| 523 |
+
"step": 168922
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 7.4,
|
| 527 |
+
"learning_rate": 6.999822329784731e-05,
|
| 528 |
+
"loss": 0.0012,
|
| 529 |
+
"step": 171236
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 7.5,
|
| 533 |
+
"learning_rate": 6.94426437328032e-05,
|
| 534 |
+
"loss": 0.0011,
|
| 535 |
+
"step": 173550
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"epoch": 7.6,
|
| 539 |
+
"learning_rate": 6.88870641677591e-05,
|
| 540 |
+
"loss": 0.0011,
|
| 541 |
+
"step": 175864
|
| 542 |
},
|
| 543 |
{
|
| 544 |
"epoch": 7.7,
|
| 545 |
+
"learning_rate": 6.8331484602715e-05,
|
| 546 |
+
"loss": 0.0012,
|
| 547 |
+
"step": 178178
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 7.8,
|
| 551 |
+
"learning_rate": 6.77759050376709e-05,
|
| 552 |
+
"loss": 0.0011,
|
| 553 |
+
"step": 180492
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 7.9,
|
| 557 |
+
"learning_rate": 6.722032547262679e-05,
|
| 558 |
+
"loss": 0.0011,
|
| 559 |
+
"step": 182806
|
| 560 |
},
|
| 561 |
{
|
| 562 |
"epoch": 8.0,
|
| 563 |
+
"eval_loss": 0.0024509974755346775,
|
| 564 |
+
"eval_max_distance": 1,
|
| 565 |
"eval_mean_distance": 0,
|
| 566 |
+
"eval_runtime": 11.9393,
|
| 567 |
+
"eval_samples_per_second": 21.609,
|
| 568 |
+
"eval_steps_per_second": 1.508,
|
| 569 |
+
"step": 185112
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 8.0,
|
| 573 |
+
"learning_rate": 6.666474590758269e-05,
|
| 574 |
+
"loss": 0.0011,
|
| 575 |
+
"step": 185120
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 8.1,
|
| 579 |
+
"learning_rate": 6.610916634253858e-05,
|
| 580 |
+
"loss": 0.0011,
|
| 581 |
+
"step": 187434
|
| 582 |
},
|
| 583 |
{
|
| 584 |
"epoch": 8.2,
|
| 585 |
+
"learning_rate": 6.555358677749447e-05,
|
| 586 |
+
"loss": 0.001,
|
| 587 |
+
"step": 189748
|
| 588 |
},
|
| 589 |
{
|
| 590 |
"epoch": 8.3,
|
| 591 |
+
"learning_rate": 6.499800721245037e-05,
|
| 592 |
+
"loss": 0.0011,
|
| 593 |
+
"step": 192062
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 8.4,
|
| 597 |
+
"learning_rate": 6.444242764740626e-05,
|
| 598 |
+
"loss": 0.0011,
|
| 599 |
+
"step": 194376
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 8.5,
|
| 603 |
+
"learning_rate": 6.388684808236215e-05,
|
| 604 |
+
"loss": 0.0011,
|
| 605 |
+
"step": 196690
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 8.6,
|
| 609 |
+
"learning_rate": 6.333126851731805e-05,
|
| 610 |
+
"loss": 0.0011,
|
| 611 |
+
"step": 199004
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 8.7,
|
| 615 |
+
"learning_rate": 6.277568895227394e-05,
|
| 616 |
+
"loss": 0.0011,
|
| 617 |
+
"step": 201318
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 8.8,
|
| 621 |
+
"learning_rate": 6.222010938722984e-05,
|
| 622 |
+
"loss": 0.0011,
|
| 623 |
+
"step": 203632
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 8.9,
|
| 627 |
+
"learning_rate": 6.166452982218573e-05,
|
| 628 |
+
"loss": 0.0011,
|
| 629 |
+
"step": 205946
|
| 630 |
},
|
| 631 |
{
|
| 632 |
"epoch": 9.0,
|
| 633 |
+
"eval_loss": 0.004092465154826641,
|
| 634 |
"eval_max_distance": 4,
|
| 635 |
"eval_mean_distance": 0,
|
| 636 |
+
"eval_runtime": 11.8563,
|
| 637 |
+
"eval_samples_per_second": 21.76,
|
| 638 |
+
"eval_steps_per_second": 1.518,
|
| 639 |
+
"step": 208251
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 9.0,
|
| 643 |
+
"learning_rate": 6.110895025714162e-05,
|
| 644 |
+
"loss": 0.0011,
|
| 645 |
+
"step": 208260
|
| 646 |
},
|
| 647 |
{
|
| 648 |
"epoch": 9.1,
|
| 649 |
+
"learning_rate": 6.055337069209752e-05,
|
| 650 |
+
"loss": 0.001,
|
| 651 |
+
"step": 210574
|
| 652 |
},
|
| 653 |
{
|
| 654 |
"epoch": 9.2,
|
| 655 |
+
"learning_rate": 5.9997791127053416e-05,
|
| 656 |
+
"loss": 0.001,
|
| 657 |
+
"step": 212888
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 9.3,
|
| 661 |
+
"learning_rate": 5.944221156200931e-05,
|
| 662 |
+
"loss": 0.001,
|
| 663 |
+
"step": 215202
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 9.4,
|
| 667 |
+
"learning_rate": 5.8886631996965203e-05,
|
| 668 |
+
"loss": 0.001,
|
| 669 |
+
"step": 217516
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 9.5,
|
| 673 |
+
"learning_rate": 5.83310524319211e-05,
|
| 674 |
+
"loss": 0.001,
|
| 675 |
+
"step": 219830
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 9.6,
|
| 679 |
+
"learning_rate": 5.777547286687699e-05,
|
| 680 |
+
"loss": 0.001,
|
| 681 |
+
"step": 222144
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 9.7,
|
| 685 |
+
"learning_rate": 5.7219893301832884e-05,
|
| 686 |
+
"loss": 0.0011,
|
| 687 |
+
"step": 224458
|
| 688 |
},
|
| 689 |
{
|
| 690 |
"epoch": 9.8,
|
| 691 |
+
"learning_rate": 5.666431373678878e-05,
|
| 692 |
+
"loss": 0.001,
|
| 693 |
+
"step": 226772
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 9.9,
|
| 697 |
+
"learning_rate": 5.610873417174468e-05,
|
| 698 |
+
"loss": 0.001,
|
| 699 |
+
"step": 229086
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 10.0,
|
| 703 |
+
"eval_loss": 0.006354992277920246,
|
| 704 |
+
"eval_max_distance": 1,
|
| 705 |
"eval_mean_distance": 0,
|
| 706 |
+
"eval_runtime": 11.9115,
|
| 707 |
+
"eval_samples_per_second": 21.66,
|
| 708 |
+
"eval_steps_per_second": 1.511,
|
| 709 |
+
"step": 231390
|
| 710 |
},
|
| 711 |
{
|
| 712 |
"epoch": 10.0,
|
| 713 |
+
"learning_rate": 5.555315460670057e-05,
|
| 714 |
+
"loss": 0.0011,
|
| 715 |
+
"step": 231400
|
| 716 |
},
|
| 717 |
{
|
| 718 |
"epoch": 10.1,
|
| 719 |
+
"learning_rate": 5.4997575041656465e-05,
|
| 720 |
+
"loss": 0.001,
|
| 721 |
+
"step": 233714
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 10.2,
|
| 725 |
+
"learning_rate": 5.444199547661236e-05,
|
| 726 |
+
"loss": 0.001,
|
| 727 |
+
"step": 236028
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 10.3,
|
| 731 |
+
"learning_rate": 5.388641591156825e-05,
|
| 732 |
+
"loss": 0.0008,
|
| 733 |
+
"step": 238342
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 10.4,
|
| 737 |
+
"learning_rate": 5.3330836346524146e-05,
|
| 738 |
+
"loss": 0.001,
|
| 739 |
+
"step": 240656
|
| 740 |
},
|
| 741 |
{
|
| 742 |
"epoch": 10.5,
|
| 743 |
+
"learning_rate": 5.277525678148004e-05,
|
| 744 |
+
"loss": 0.001,
|
| 745 |
+
"step": 242970
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 10.6,
|
| 749 |
+
"learning_rate": 5.2219677216435933e-05,
|
| 750 |
+
"loss": 0.0009,
|
| 751 |
+
"step": 245284
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 10.7,
|
| 755 |
+
"learning_rate": 5.166409765139183e-05,
|
| 756 |
+
"loss": 0.001,
|
| 757 |
+
"step": 247598
|
| 758 |
},
|
| 759 |
{
|
| 760 |
"epoch": 10.8,
|
| 761 |
+
"learning_rate": 5.110851808634772e-05,
|
| 762 |
+
"loss": 0.001,
|
| 763 |
+
"step": 249912
|
| 764 |
},
|
| 765 |
{
|
| 766 |
"epoch": 10.9,
|
| 767 |
+
"learning_rate": 5.0552938521303614e-05,
|
| 768 |
+
"loss": 0.001,
|
| 769 |
+
"step": 252226
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 11.0,
|
| 773 |
+
"eval_loss": 0.0059073311276733875,
|
| 774 |
+
"eval_max_distance": 1,
|
| 775 |
"eval_mean_distance": 0,
|
| 776 |
+
"eval_runtime": 11.9148,
|
| 777 |
+
"eval_samples_per_second": 21.654,
|
| 778 |
+
"eval_steps_per_second": 1.511,
|
| 779 |
+
"step": 254529
|
| 780 |
},
|
| 781 |
{
|
| 782 |
"epoch": 11.0,
|
| 783 |
+
"learning_rate": 4.9997358956259515e-05,
|
| 784 |
+
"loss": 0.001,
|
| 785 |
+
"step": 254540
|
| 786 |
},
|
| 787 |
{
|
| 788 |
"epoch": 11.1,
|
| 789 |
+
"learning_rate": 4.944177939121541e-05,
|
| 790 |
+
"loss": 0.0009,
|
| 791 |
+
"step": 256854
|
| 792 |
},
|
| 793 |
{
|
| 794 |
"epoch": 11.2,
|
| 795 |
+
"learning_rate": 4.88861998261713e-05,
|
| 796 |
+
"loss": 0.0009,
|
| 797 |
+
"step": 259168
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 11.3,
|
| 801 |
+
"learning_rate": 4.83306202611272e-05,
|
| 802 |
+
"loss": 0.0009,
|
| 803 |
+
"step": 261482
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 11.4,
|
| 807 |
+
"learning_rate": 4.7775040696083096e-05,
|
| 808 |
+
"loss": 0.0009,
|
| 809 |
+
"step": 263796
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 11.5,
|
| 813 |
+
"learning_rate": 4.721946113103899e-05,
|
| 814 |
+
"loss": 0.001,
|
| 815 |
+
"step": 266110
|
| 816 |
},
|
| 817 |
{
|
| 818 |
"epoch": 11.6,
|
| 819 |
+
"learning_rate": 4.666388156599488e-05,
|
| 820 |
+
"loss": 0.0009,
|
| 821 |
+
"step": 268424
|
| 822 |
},
|
| 823 |
{
|
| 824 |
"epoch": 11.7,
|
| 825 |
+
"learning_rate": 4.610830200095078e-05,
|
| 826 |
+
"loss": 0.0009,
|
| 827 |
+
"step": 270738
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 11.8,
|
| 831 |
+
"learning_rate": 4.555272243590668e-05,
|
| 832 |
+
"loss": 0.0009,
|
| 833 |
+
"step": 273052
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 11.9,
|
| 837 |
+
"learning_rate": 4.499714287086257e-05,
|
| 838 |
+
"loss": 0.0008,
|
| 839 |
+
"step": 275366
|
| 840 |
},
|
| 841 |
{
|
| 842 |
"epoch": 12.0,
|
| 843 |
+
"eval_loss": 0.004573535639792681,
|
| 844 |
"eval_max_distance": 3,
|
| 845 |
"eval_mean_distance": 0,
|
| 846 |
+
"eval_runtime": 11.88,
|
| 847 |
+
"eval_samples_per_second": 21.717,
|
| 848 |
+
"eval_steps_per_second": 1.515,
|
| 849 |
+
"step": 277668
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 12.0,
|
| 853 |
+
"learning_rate": 4.4441563305818464e-05,
|
| 854 |
+
"loss": 0.001,
|
| 855 |
+
"step": 277680
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 12.1,
|
| 859 |
+
"learning_rate": 4.388598374077436e-05,
|
| 860 |
+
"loss": 0.0008,
|
| 861 |
+
"step": 279994
|
| 862 |
},
|
| 863 |
{
|
| 864 |
"epoch": 12.2,
|
| 865 |
+
"learning_rate": 4.333040417573025e-05,
|
| 866 |
+
"loss": 0.0009,
|
| 867 |
+
"step": 282308
|
| 868 |
},
|
| 869 |
{
|
| 870 |
"epoch": 12.3,
|
| 871 |
+
"learning_rate": 4.2774824610686145e-05,
|
| 872 |
+
"loss": 0.001,
|
| 873 |
+
"step": 284622
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 12.4,
|
| 877 |
+
"learning_rate": 4.221924504564204e-05,
|
| 878 |
+
"loss": 0.0008,
|
| 879 |
+
"step": 286936
|
| 880 |
},
|
| 881 |
{
|
| 882 |
"epoch": 12.5,
|
| 883 |
+
"learning_rate": 4.166366548059793e-05,
|
| 884 |
+
"loss": 0.0009,
|
| 885 |
+
"step": 289250
|
| 886 |
},
|
| 887 |
{
|
| 888 |
"epoch": 12.6,
|
| 889 |
+
"learning_rate": 4.1108085915553826e-05,
|
| 890 |
+
"loss": 0.0008,
|
| 891 |
+
"step": 291564
|
| 892 |
},
|
| 893 |
{
|
| 894 |
"epoch": 12.7,
|
| 895 |
+
"learning_rate": 4.055250635050972e-05,
|
| 896 |
+
"loss": 0.0009,
|
| 897 |
+
"step": 293878
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 12.8,
|
| 901 |
+
"learning_rate": 3.999692678546561e-05,
|
| 902 |
+
"loss": 0.0008,
|
| 903 |
+
"step": 296192
|
| 904 |
},
|
| 905 |
{
|
| 906 |
"epoch": 12.9,
|
| 907 |
+
"learning_rate": 3.9441347220421513e-05,
|
| 908 |
+
"loss": 0.0008,
|
| 909 |
+
"step": 298506
|
| 910 |
},
|
| 911 |
{
|
| 912 |
"epoch": 13.0,
|
| 913 |
+
"eval_loss": 0.004101805854588747,
|
| 914 |
+
"eval_max_distance": 2,
|
| 915 |
"eval_mean_distance": 0,
|
| 916 |
+
"eval_runtime": 11.92,
|
| 917 |
+
"eval_samples_per_second": 21.644,
|
| 918 |
+
"eval_steps_per_second": 1.51,
|
| 919 |
+
"step": 300807
|
| 920 |
},
|
| 921 |
{
|
| 922 |
"epoch": 13.0,
|
| 923 |
+
"learning_rate": 3.888576765537741e-05,
|
| 924 |
+
"loss": 0.0009,
|
| 925 |
+
"step": 300820
|
| 926 |
},
|
| 927 |
{
|
| 928 |
"epoch": 13.1,
|
| 929 |
+
"learning_rate": 3.83301880903333e-05,
|
| 930 |
+
"loss": 0.0008,
|
| 931 |
+
"step": 303134
|
| 932 |
},
|
| 933 |
{
|
| 934 |
"epoch": 13.2,
|
| 935 |
+
"learning_rate": 3.7774608525289194e-05,
|
| 936 |
+
"loss": 0.0008,
|
| 937 |
+
"step": 305448
|
| 938 |
},
|
| 939 |
{
|
| 940 |
"epoch": 13.3,
|
| 941 |
+
"learning_rate": 3.721902896024509e-05,
|
| 942 |
+
"loss": 0.0008,
|
| 943 |
+
"step": 307762
|
| 944 |
},
|
| 945 |
{
|
| 946 |
"epoch": 13.4,
|
| 947 |
+
"learning_rate": 3.666344939520098e-05,
|
| 948 |
+
"loss": 0.0009,
|
| 949 |
+
"step": 310076
|
| 950 |
},
|
| 951 |
{
|
| 952 |
"epoch": 13.5,
|
| 953 |
+
"learning_rate": 3.610786983015688e-05,
|
| 954 |
+
"loss": 0.0009,
|
| 955 |
+
"step": 312390
|
| 956 |
},
|
| 957 |
{
|
| 958 |
"epoch": 13.6,
|
| 959 |
+
"learning_rate": 3.5552290265112775e-05,
|
| 960 |
+
"loss": 0.0008,
|
| 961 |
+
"step": 314704
|
| 962 |
},
|
| 963 |
{
|
| 964 |
"epoch": 13.7,
|
| 965 |
+
"learning_rate": 3.499671070006867e-05,
|
| 966 |
+
"loss": 0.0009,
|
| 967 |
+
"step": 317018
|
| 968 |
},
|
| 969 |
{
|
| 970 |
"epoch": 13.8,
|
| 971 |
+
"learning_rate": 3.444113113502456e-05,
|
| 972 |
+
"loss": 0.0009,
|
| 973 |
+
"step": 319332
|
| 974 |
},
|
| 975 |
{
|
| 976 |
"epoch": 13.9,
|
| 977 |
+
"learning_rate": 3.3885551569980456e-05,
|
| 978 |
+
"loss": 0.0008,
|
| 979 |
+
"step": 321646
|
| 980 |
},
|
| 981 |
{
|
| 982 |
"epoch": 14.0,
|
| 983 |
+
"eval_loss": 0.005928508937358856,
|
| 984 |
+
"eval_max_distance": 2,
|
| 985 |
"eval_mean_distance": 0,
|
| 986 |
+
"eval_runtime": 12.049,
|
| 987 |
+
"eval_samples_per_second": 21.413,
|
| 988 |
+
"eval_steps_per_second": 1.494,
|
| 989 |
+
"step": 323946
|
| 990 |
},
|
| 991 |
{
|
| 992 |
"epoch": 14.0,
|
| 993 |
+
"learning_rate": 3.332997200493636e-05,
|
| 994 |
+
"loss": 0.0009,
|
| 995 |
+
"step": 323960
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 14.1,
|
| 999 |
+
"learning_rate": 3.277439243989225e-05,
|
| 1000 |
+
"loss": 0.0008,
|
| 1001 |
+
"step": 326274
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 14.2,
|
| 1005 |
+
"learning_rate": 3.2218812874848144e-05,
|
| 1006 |
+
"loss": 0.0008,
|
| 1007 |
+
"step": 328588
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 14.3,
|
| 1011 |
+
"learning_rate": 3.166323330980404e-05,
|
| 1012 |
+
"loss": 0.0007,
|
| 1013 |
+
"step": 330902
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 14.4,
|
| 1017 |
+
"learning_rate": 3.110765374475993e-05,
|
| 1018 |
+
"loss": 0.0008,
|
| 1019 |
+
"step": 333216
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 14.5,
|
| 1023 |
+
"learning_rate": 3.0552074179715825e-05,
|
| 1024 |
+
"loss": 0.0007,
|
| 1025 |
+
"step": 335530
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 14.6,
|
| 1029 |
+
"learning_rate": 2.9996494614671722e-05,
|
| 1030 |
+
"loss": 0.0008,
|
| 1031 |
+
"step": 337844
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 14.7,
|
| 1035 |
+
"learning_rate": 2.9440915049627615e-05,
|
| 1036 |
+
"loss": 0.0008,
|
| 1037 |
+
"step": 340158
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 14.8,
|
| 1041 |
+
"learning_rate": 2.888533548458351e-05,
|
| 1042 |
+
"loss": 0.0008,
|
| 1043 |
+
"step": 342472
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 14.9,
|
| 1047 |
+
"learning_rate": 2.8329755919539403e-05,
|
| 1048 |
+
"loss": 0.0008,
|
| 1049 |
+
"step": 344786
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 15.0,
|
| 1053 |
+
"eval_loss": 0.005999278277158737,
|
| 1054 |
+
"eval_max_distance": 2,
|
| 1055 |
+
"eval_mean_distance": 0,
|
| 1056 |
+
"eval_runtime": 11.9346,
|
| 1057 |
+
"eval_samples_per_second": 21.618,
|
| 1058 |
+
"eval_steps_per_second": 1.508,
|
| 1059 |
+
"step": 347085
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 15.0,
|
| 1063 |
+
"learning_rate": 2.7774176354495296e-05,
|
| 1064 |
+
"loss": 0.0008,
|
| 1065 |
+
"step": 347100
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 15.1,
|
| 1069 |
+
"learning_rate": 2.7218596789451197e-05,
|
| 1070 |
+
"loss": 0.0007,
|
| 1071 |
+
"step": 349414
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 15.2,
|
| 1075 |
+
"learning_rate": 2.666301722440709e-05,
|
| 1076 |
+
"loss": 0.0007,
|
| 1077 |
+
"step": 351728
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 15.3,
|
| 1081 |
+
"learning_rate": 2.6107437659362984e-05,
|
| 1082 |
+
"loss": 0.0007,
|
| 1083 |
+
"step": 354042
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 15.4,
|
| 1087 |
+
"learning_rate": 2.5551858094318877e-05,
|
| 1088 |
+
"loss": 0.0007,
|
| 1089 |
+
"step": 356356
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 15.5,
|
| 1093 |
+
"learning_rate": 2.4996278529274768e-05,
|
| 1094 |
+
"loss": 0.0007,
|
| 1095 |
+
"step": 358670
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 15.6,
|
| 1099 |
+
"learning_rate": 2.4440698964230665e-05,
|
| 1100 |
+
"loss": 0.0007,
|
| 1101 |
+
"step": 360984
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 15.7,
|
| 1105 |
+
"learning_rate": 2.3885119399186558e-05,
|
| 1106 |
+
"loss": 0.0007,
|
| 1107 |
+
"step": 363298
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 15.8,
|
| 1111 |
+
"learning_rate": 2.3329539834142455e-05,
|
| 1112 |
+
"loss": 0.0008,
|
| 1113 |
+
"step": 365612
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 15.9,
|
| 1117 |
+
"learning_rate": 2.277396026909835e-05,
|
| 1118 |
+
"loss": 0.0007,
|
| 1119 |
+
"step": 367926
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 16.0,
|
| 1123 |
+
"eval_loss": 0.005278590135276318,
|
| 1124 |
+
"eval_max_distance": 2,
|
| 1125 |
+
"eval_mean_distance": 0,
|
| 1126 |
+
"eval_runtime": 11.8758,
|
| 1127 |
+
"eval_samples_per_second": 21.725,
|
| 1128 |
+
"eval_steps_per_second": 1.516,
|
| 1129 |
+
"step": 370224
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 16.0,
|
| 1133 |
+
"learning_rate": 2.2218380704054242e-05,
|
| 1134 |
+
"loss": 0.0008,
|
| 1135 |
+
"step": 370240
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 16.1,
|
| 1139 |
+
"learning_rate": 2.166280113901014e-05,
|
| 1140 |
+
"loss": 0.0007,
|
| 1141 |
+
"step": 372554
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 16.2,
|
| 1145 |
+
"learning_rate": 2.1107221573966033e-05,
|
| 1146 |
+
"loss": 0.0007,
|
| 1147 |
+
"step": 374868
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 16.3,
|
| 1151 |
+
"learning_rate": 2.0551642008921927e-05,
|
| 1152 |
+
"loss": 0.0007,
|
| 1153 |
+
"step": 377182
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 16.4,
|
| 1157 |
+
"learning_rate": 1.9996062443877824e-05,
|
| 1158 |
+
"loss": 0.0007,
|
| 1159 |
+
"step": 379496
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 16.5,
|
| 1163 |
+
"learning_rate": 1.9440482878833717e-05,
|
| 1164 |
+
"loss": 0.0007,
|
| 1165 |
+
"step": 381810
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 16.6,
|
| 1169 |
+
"learning_rate": 1.888490331378961e-05,
|
| 1170 |
+
"loss": 0.0007,
|
| 1171 |
+
"step": 384124
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 16.7,
|
| 1175 |
+
"learning_rate": 1.8329323748745504e-05,
|
| 1176 |
+
"loss": 0.0008,
|
| 1177 |
+
"step": 386438
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 16.8,
|
| 1181 |
+
"learning_rate": 1.7773744183701398e-05,
|
| 1182 |
+
"loss": 0.0007,
|
| 1183 |
+
"step": 388752
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 16.9,
|
| 1187 |
+
"learning_rate": 1.7218164618657295e-05,
|
| 1188 |
+
"loss": 0.0007,
|
| 1189 |
+
"step": 391066
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 17.0,
|
| 1193 |
+
"eval_loss": 0.004050484858453274,
|
| 1194 |
+
"eval_max_distance": 1,
|
| 1195 |
+
"eval_mean_distance": 0,
|
| 1196 |
+
"eval_runtime": 11.8948,
|
| 1197 |
+
"eval_samples_per_second": 21.69,
|
| 1198 |
+
"eval_steps_per_second": 1.513,
|
| 1199 |
+
"step": 393363
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 17.0,
|
| 1203 |
+
"learning_rate": 1.666258505361319e-05,
|
| 1204 |
+
"loss": 0.0007,
|
| 1205 |
+
"step": 393380
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 17.1,
|
| 1209 |
+
"learning_rate": 1.6107005488569082e-05,
|
| 1210 |
+
"loss": 0.0007,
|
| 1211 |
+
"step": 395694
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 17.2,
|
| 1215 |
+
"learning_rate": 1.555142592352498e-05,
|
| 1216 |
+
"loss": 0.0007,
|
| 1217 |
+
"step": 398008
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 17.3,
|
| 1221 |
+
"learning_rate": 1.4995846358480873e-05,
|
| 1222 |
+
"loss": 0.0006,
|
| 1223 |
+
"step": 400322
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 17.4,
|
| 1227 |
+
"learning_rate": 1.4440266793436766e-05,
|
| 1228 |
+
"loss": 0.0007,
|
| 1229 |
+
"step": 402636
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 17.5,
|
| 1233 |
+
"learning_rate": 1.3884687228392662e-05,
|
| 1234 |
+
"loss": 0.0007,
|
| 1235 |
+
"step": 404950
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 17.6,
|
| 1239 |
+
"learning_rate": 1.3329107663348555e-05,
|
| 1240 |
+
"loss": 0.0007,
|
| 1241 |
+
"step": 407264
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 17.7,
|
| 1245 |
+
"learning_rate": 1.2773528098304449e-05,
|
| 1246 |
+
"loss": 0.0007,
|
| 1247 |
+
"step": 409578
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 17.8,
|
| 1251 |
+
"learning_rate": 1.2217948533260346e-05,
|
| 1252 |
+
"loss": 0.0007,
|
| 1253 |
+
"step": 411892
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 17.9,
|
| 1257 |
+
"learning_rate": 1.166236896821624e-05,
|
| 1258 |
+
"loss": 0.0007,
|
| 1259 |
+
"step": 414206
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 18.0,
|
| 1263 |
+
"eval_loss": 0.004916150122880936,
|
| 1264 |
+
"eval_max_distance": 1,
|
| 1265 |
+
"eval_mean_distance": 0,
|
| 1266 |
+
"eval_runtime": 11.9279,
|
| 1267 |
+
"eval_samples_per_second": 21.63,
|
| 1268 |
+
"eval_steps_per_second": 1.509,
|
| 1269 |
+
"step": 416502
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 18.0,
|
| 1273 |
+
"learning_rate": 1.1106789403172135e-05,
|
| 1274 |
+
"loss": 0.0007,
|
| 1275 |
+
"step": 416520
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 18.1,
|
| 1279 |
+
"learning_rate": 1.0551209838128028e-05,
|
| 1280 |
+
"loss": 0.0007,
|
| 1281 |
+
"step": 418834
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 18.2,
|
| 1285 |
+
"learning_rate": 9.995630273083924e-06,
|
| 1286 |
+
"loss": 0.0007,
|
| 1287 |
+
"step": 421148
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 18.3,
|
| 1291 |
+
"learning_rate": 9.440050708039817e-06,
|
| 1292 |
+
"loss": 0.0006,
|
| 1293 |
+
"step": 423462
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 18.4,
|
| 1297 |
+
"learning_rate": 8.884471142995713e-06,
|
| 1298 |
+
"loss": 0.0006,
|
| 1299 |
+
"step": 425776
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 18.5,
|
| 1303 |
+
"learning_rate": 8.328891577951608e-06,
|
| 1304 |
+
"loss": 0.0006,
|
| 1305 |
+
"step": 428090
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 18.6,
|
| 1309 |
+
"learning_rate": 7.773312012907501e-06,
|
| 1310 |
+
"loss": 0.0007,
|
| 1311 |
+
"step": 430404
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 18.7,
|
| 1315 |
+
"learning_rate": 7.217732447863396e-06,
|
| 1316 |
+
"loss": 0.0006,
|
| 1317 |
+
"step": 432718
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 18.8,
|
| 1321 |
+
"learning_rate": 6.66215288281929e-06,
|
| 1322 |
+
"loss": 0.0007,
|
| 1323 |
+
"step": 435032
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 18.9,
|
| 1327 |
+
"learning_rate": 6.106573317775185e-06,
|
| 1328 |
+
"loss": 0.0007,
|
| 1329 |
+
"step": 437346
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 19.0,
|
| 1333 |
+
"eval_loss": 0.005351274739950895,
|
| 1334 |
+
"eval_max_distance": 1,
|
| 1335 |
+
"eval_mean_distance": 0,
|
| 1336 |
+
"eval_runtime": 11.9441,
|
| 1337 |
+
"eval_samples_per_second": 21.601,
|
| 1338 |
+
"eval_steps_per_second": 1.507,
|
| 1339 |
+
"step": 439641
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 19.0,
|
| 1343 |
+
"learning_rate": 5.550993752731079e-06,
|
| 1344 |
+
"loss": 0.0007,
|
| 1345 |
+
"step": 439660
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 19.1,
|
| 1349 |
+
"learning_rate": 4.995414187686974e-06,
|
| 1350 |
+
"loss": 0.0006,
|
| 1351 |
+
"step": 441974
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 19.2,
|
| 1355 |
+
"learning_rate": 4.439834622642869e-06,
|
| 1356 |
+
"loss": 0.0007,
|
| 1357 |
+
"step": 444288
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 19.3,
|
| 1361 |
+
"learning_rate": 3.8842550575987635e-06,
|
| 1362 |
+
"loss": 0.0007,
|
| 1363 |
+
"step": 446602
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 19.4,
|
| 1367 |
+
"learning_rate": 3.328675492554658e-06,
|
| 1368 |
+
"loss": 0.0007,
|
| 1369 |
+
"step": 448916
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 19.5,
|
| 1373 |
+
"learning_rate": 2.7730959275105524e-06,
|
| 1374 |
+
"loss": 0.0006,
|
| 1375 |
+
"step": 451230
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 19.6,
|
| 1379 |
+
"learning_rate": 2.217516362466447e-06,
|
| 1380 |
+
"loss": 0.0006,
|
| 1381 |
+
"step": 453544
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 19.7,
|
| 1385 |
+
"learning_rate": 1.6619367974223413e-06,
|
| 1386 |
+
"loss": 0.0006,
|
| 1387 |
+
"step": 455858
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 19.8,
|
| 1391 |
+
"learning_rate": 1.106357232378236e-06,
|
| 1392 |
+
"loss": 0.0006,
|
| 1393 |
+
"step": 458172
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 19.9,
|
| 1397 |
+
"learning_rate": 5.507776673341304e-07,
|
| 1398 |
+
"loss": 0.0006,
|
| 1399 |
+
"step": 460486
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 20.0,
|
| 1403 |
+
"eval_loss": 0.005220064427703619,
|
| 1404 |
+
"eval_max_distance": 1,
|
| 1405 |
+
"eval_mean_distance": 0,
|
| 1406 |
+
"eval_runtime": 12.038,
|
| 1407 |
+
"eval_samples_per_second": 21.432,
|
| 1408 |
+
"eval_steps_per_second": 1.495,
|
| 1409 |
+
"step": 462780
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 20.0,
|
| 1413 |
+
"step": 462780,
|
| 1414 |
+
"total_flos": 1.1723128343387136e+17,
|
| 1415 |
+
"train_loss": 0.000996548650895231,
|
| 1416 |
+
"train_runtime": 32132.3816,
|
| 1417 |
+
"train_samples_per_second": 216.028,
|
| 1418 |
+
"train_steps_per_second": 14.402
|
| 1419 |
}
|
| 1420 |
],
|
| 1421 |
+
"logging_steps": 2314,
|
| 1422 |
+
"max_steps": 462780,
|
| 1423 |
+
"num_train_epochs": 20,
|
| 1424 |
+
"save_steps": 4628,
|
| 1425 |
+
"total_flos": 1.1723128343387136e+17,
|
| 1426 |
"trial_name": null,
|
| 1427 |
"trial_params": null
|
| 1428 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4091
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:828b4d601bb77ebe23d0f70b8f8d6a83d87ce2df6d10f535b88504dd103d0ee2
|
| 3 |
size 4091
|