moussaKam
commited on
Commit
•
18881bd
1
Parent(s):
ca6d28c
fix model
Browse files- all_results.json +10 -2
- config.json +1 -1
- pytorch_model.bin +1 -1
- tokenizer_config.json +1 -1
- train_results.json +10 -2
- trainer_state.json +118 -346
- training_args.bin +2 -2
all_results.json
CHANGED
|
@@ -1,6 +1,14 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
"train_samples": 200000,
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"init_mem_cpu_alloc_delta": 2169131008,
|
| 4 |
+
"init_mem_cpu_peaked_delta": 9641984,
|
| 5 |
+
"init_mem_gpu_alloc_delta": 17548800,
|
| 6 |
+
"init_mem_gpu_peaked_delta": 0,
|
| 7 |
+
"train_mem_cpu_alloc_delta": 950534144,
|
| 8 |
+
"train_mem_cpu_peaked_delta": 4513792,
|
| 9 |
+
"train_mem_gpu_alloc_delta": 53029376,
|
| 10 |
+
"train_mem_gpu_peaked_delta": 669877760,
|
| 11 |
+
"train_runtime": 1200.1754,
|
| 12 |
"train_samples": 200000,
|
| 13 |
+
"train_samples_per_second": 15.623
|
| 14 |
}
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "checkpoints/tiny_bert",
|
| 3 |
"architectures": [
|
| 4 |
"BertForSequenceClassification"
|
| 5 |
],
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "checkpoints/pretraining_checkpoints/translation_dataset/tiny_bert/",
|
| 3 |
"architectures": [
|
| 4 |
"BertForSequenceClassification"
|
| 5 |
],
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 17565609
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4313aa6b0cabffea442f2107bc0904af3ca4c37a9d4ad7b49ab5ae25f8d3b77
|
| 3 |
size 17565609
|
tokenizer_config.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "name_or_path": "checkpoints/tiny_bert", "do_basic_tokenize": true, "never_split": null}
|
|
|
|
| 1 |
+
{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "name_or_path": "checkpoints/pretraining_checkpoints/translation_dataset/tiny_bert/", "do_basic_tokenize": true, "never_split": null}
|
train_results.json
CHANGED
|
@@ -1,6 +1,14 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
"train_samples": 200000,
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"init_mem_cpu_alloc_delta": 2169131008,
|
| 4 |
+
"init_mem_cpu_peaked_delta": 9641984,
|
| 5 |
+
"init_mem_gpu_alloc_delta": 17548800,
|
| 6 |
+
"init_mem_gpu_peaked_delta": 0,
|
| 7 |
+
"train_mem_cpu_alloc_delta": 950534144,
|
| 8 |
+
"train_mem_cpu_peaked_delta": 4513792,
|
| 9 |
+
"train_mem_gpu_alloc_delta": 53029376,
|
| 10 |
+
"train_mem_gpu_peaked_delta": 669877760,
|
| 11 |
+
"train_runtime": 1200.1754,
|
| 12 |
"train_samples": 200000,
|
| 13 |
+
"train_samples_per_second": 15.623
|
| 14 |
}
|
trainer_state.json
CHANGED
|
@@ -2,472 +2,244 @@
|
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 3.0,
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
-
{
|
| 11 |
-
"epoch": 0.04,
|
| 12 |
-
"learning_rate": 4.4355555555555555e-06,
|
| 13 |
-
"loss": 0.0959,
|
| 14 |
-
"step": 500
|
| 15 |
-
},
|
| 16 |
{
|
| 17 |
"epoch": 0.08,
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 0.
|
| 20 |
-
"step":
|
| 21 |
-
},
|
| 22 |
-
{
|
| 23 |
-
"epoch": 0.12,
|
| 24 |
-
"learning_rate": 1.3324444444444446e-05,
|
| 25 |
-
"loss": 0.0042,
|
| 26 |
-
"step": 1500
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.16,
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 0.
|
| 32 |
-
"step":
|
| 33 |
-
},
|
| 34 |
-
{
|
| 35 |
-
"epoch": 0.2,
|
| 36 |
-
"learning_rate": 1.985872340425532e-05,
|
| 37 |
-
"loss": 0.0033,
|
| 38 |
-
"step": 2500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.24,
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
-
"step":
|
| 45 |
-
},
|
| 46 |
-
{
|
| 47 |
-
"epoch": 0.28,
|
| 48 |
-
"learning_rate": 1.92913475177305e-05,
|
| 49 |
-
"loss": 0.0031,
|
| 50 |
-
"step": 3500
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.32,
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 0.
|
| 56 |
-
"step":
|
| 57 |
-
},
|
| 58 |
-
{
|
| 59 |
-
"epoch": 0.36,
|
| 60 |
-
"learning_rate": 1.8723971631205676e-05,
|
| 61 |
-
"loss": 0.0031,
|
| 62 |
-
"step": 4500
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.4,
|
| 66 |
-
"learning_rate":
|
| 67 |
-
"loss": 0.
|
| 68 |
-
"step":
|
| 69 |
-
},
|
| 70 |
-
{
|
| 71 |
-
"epoch": 0.44,
|
| 72 |
-
"learning_rate": 1.815659574468085e-05,
|
| 73 |
-
"loss": 0.0029,
|
| 74 |
-
"step": 5500
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.48,
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 0.
|
| 80 |
-
"step":
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.52,
|
| 84 |
-
"learning_rate": 1.758921985815603e-05,
|
| 85 |
-
"loss": 0.0029,
|
| 86 |
-
"step": 6500
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.56,
|
| 90 |
-
"learning_rate":
|
| 91 |
-
"loss": 0.
|
| 92 |
-
"step":
|
| 93 |
-
},
|
| 94 |
-
{
|
| 95 |
-
"epoch": 0.6,
|
| 96 |
-
"learning_rate": 1.7021843971631207e-05,
|
| 97 |
-
"loss": 0.0029,
|
| 98 |
-
"step": 7500
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 0.64,
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss": 0.
|
| 104 |
-
"step":
|
| 105 |
-
},
|
| 106 |
-
{
|
| 107 |
-
"epoch": 0.68,
|
| 108 |
-
"learning_rate": 1.6454468085106386e-05,
|
| 109 |
-
"loss": 0.0028,
|
| 110 |
-
"step": 8500
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 0.72,
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss": 0.
|
| 116 |
-
"step":
|
| 117 |
-
},
|
| 118 |
-
{
|
| 119 |
-
"epoch": 0.76,
|
| 120 |
-
"learning_rate": 1.5887092198581562e-05,
|
| 121 |
-
"loss": 0.0028,
|
| 122 |
-
"step": 9500
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.8,
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 0.
|
| 128 |
-
"step":
|
| 129 |
-
},
|
| 130 |
-
{
|
| 131 |
-
"epoch": 0.84,
|
| 132 |
-
"learning_rate": 1.5320283687943263e-05,
|
| 133 |
-
"loss": 0.0028,
|
| 134 |
-
"step": 10500
|
| 135 |
},
|
| 136 |
{
|
| 137 |
"epoch": 0.88,
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"loss": 0.
|
| 140 |
-
"step":
|
| 141 |
-
},
|
| 142 |
-
{
|
| 143 |
-
"epoch": 0.92,
|
| 144 |
-
"learning_rate": 1.4753475177304965e-05,
|
| 145 |
-
"loss": 0.0028,
|
| 146 |
-
"step": 11500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 0.96,
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss": 0.
|
| 152 |
-
"step":
|
| 153 |
-
},
|
| 154 |
-
{
|
| 155 |
-
"epoch": 1.0,
|
| 156 |
-
"learning_rate": 1.4186099290780141e-05,
|
| 157 |
-
"loss": 0.0028,
|
| 158 |
-
"step": 12500
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 1.04,
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 0.
|
| 164 |
-
"step":
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 1.08,
|
| 168 |
-
"learning_rate": 1.361872340425532e-05,
|
| 169 |
-
"loss": 0.0026,
|
| 170 |
-
"step": 13500
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 1.12,
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 0.
|
| 176 |
-
"step":
|
| 177 |
-
},
|
| 178 |
-
{
|
| 179 |
-
"epoch": 1.16,
|
| 180 |
-
"learning_rate": 1.3051347517730497e-05,
|
| 181 |
-
"loss": 0.0026,
|
| 182 |
-
"step": 14500
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 1.2,
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
-
"step":
|
| 189 |
-
},
|
| 190 |
-
{
|
| 191 |
-
"epoch": 1.24,
|
| 192 |
-
"learning_rate": 1.24845390070922e-05,
|
| 193 |
-
"loss": 0.0026,
|
| 194 |
-
"step": 15500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 1.28,
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 0.
|
| 200 |
-
"step":
|
| 201 |
-
},
|
| 202 |
-
{
|
| 203 |
-
"epoch": 1.32,
|
| 204 |
-
"learning_rate": 1.1917730496453901e-05,
|
| 205 |
-
"loss": 0.0026,
|
| 206 |
-
"step": 16500
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 1.36,
|
| 210 |
-
"learning_rate":
|
| 211 |
-
"loss": 0.
|
| 212 |
-
"step":
|
| 213 |
-
},
|
| 214 |
-
{
|
| 215 |
-
"epoch": 1.4,
|
| 216 |
-
"learning_rate": 1.1350354609929079e-05,
|
| 217 |
-
"loss": 0.0027,
|
| 218 |
-
"step": 17500
|
| 219 |
},
|
| 220 |
{
|
| 221 |
"epoch": 1.44,
|
| 222 |
-
"learning_rate":
|
| 223 |
-
"loss": 0.
|
| 224 |
-
"step":
|
| 225 |
-
},
|
| 226 |
-
{
|
| 227 |
-
"epoch": 1.48,
|
| 228 |
-
"learning_rate": 1.0782978723404257e-05,
|
| 229 |
-
"loss": 0.0027,
|
| 230 |
-
"step": 18500
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 1.52,
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss": 0.
|
| 236 |
-
"step":
|
| 237 |
-
},
|
| 238 |
-
{
|
| 239 |
-
"epoch": 1.56,
|
| 240 |
-
"learning_rate": 1.0215602836879433e-05,
|
| 241 |
-
"loss": 0.0026,
|
| 242 |
-
"step": 19500
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 1.6,
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
-
"step":
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 1.64,
|
| 252 |
-
"learning_rate": 9.648794326241135e-06,
|
| 253 |
-
"loss": 0.0026,
|
| 254 |
-
"step": 20500
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 1.68,
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
-
"step":
|
| 261 |
-
},
|
| 262 |
-
{
|
| 263 |
-
"epoch": 1.72,
|
| 264 |
-
"learning_rate": 9.081418439716313e-06,
|
| 265 |
-
"loss": 0.0025,
|
| 266 |
-
"step": 21500
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 1.76,
|
| 270 |
-
"learning_rate":
|
| 271 |
-
"loss": 0.
|
| 272 |
-
"step":
|
| 273 |
-
},
|
| 274 |
-
{
|
| 275 |
-
"epoch": 1.8,
|
| 276 |
-
"learning_rate": 8.514609929078015e-06,
|
| 277 |
-
"loss": 0.0025,
|
| 278 |
-
"step": 22500
|
| 279 |
},
|
| 280 |
{
|
| 281 |
"epoch": 1.84,
|
| 282 |
-
"learning_rate":
|
| 283 |
-
"loss": 0.
|
| 284 |
-
"step":
|
| 285 |
-
},
|
| 286 |
-
{
|
| 287 |
-
"epoch": 1.88,
|
| 288 |
-
"learning_rate": 7.947234042553192e-06,
|
| 289 |
-
"loss": 0.0025,
|
| 290 |
-
"step": 23500
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 1.92,
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss": 0.
|
| 296 |
-
"step":
|
| 297 |
-
},
|
| 298 |
-
{
|
| 299 |
-
"epoch": 1.96,
|
| 300 |
-
"learning_rate": 7.380425531914894e-06,
|
| 301 |
-
"loss": 0.0026,
|
| 302 |
-
"step": 24500
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 2.0,
|
| 306 |
-
"learning_rate":
|
| 307 |
-
"loss": 0.
|
| 308 |
-
"step":
|
| 309 |
-
},
|
| 310 |
-
{
|
| 311 |
-
"epoch": 2.04,
|
| 312 |
-
"learning_rate": 6.813049645390072e-06,
|
| 313 |
-
"loss": 0.0024,
|
| 314 |
-
"step": 25500
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 2.08,
|
| 318 |
-
"learning_rate":
|
| 319 |
-
"loss": 0.
|
| 320 |
-
"step":
|
| 321 |
-
},
|
| 322 |
-
{
|
| 323 |
-
"epoch": 2.12,
|
| 324 |
-
"learning_rate": 6.246241134751774e-06,
|
| 325 |
-
"loss": 0.0024,
|
| 326 |
-
"step": 26500
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 2.16,
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 0.
|
| 332 |
-
"step":
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 2.2,
|
| 336 |
-
"learning_rate": 5.6788652482269516e-06,
|
| 337 |
-
"loss": 0.0024,
|
| 338 |
-
"step": 27500
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 2.24,
|
| 342 |
-
"learning_rate":
|
| 343 |
-
"loss": 0.
|
| 344 |
-
"step":
|
| 345 |
-
},
|
| 346 |
-
{
|
| 347 |
-
"epoch": 2.28,
|
| 348 |
-
"learning_rate": 5.112056737588653e-06,
|
| 349 |
-
"loss": 0.0024,
|
| 350 |
-
"step": 28500
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 2.32,
|
| 354 |
-
"learning_rate":
|
| 355 |
-
"loss": 0.
|
| 356 |
-
"step":
|
| 357 |
-
},
|
| 358 |
-
{
|
| 359 |
-
"epoch": 2.36,
|
| 360 |
-
"learning_rate": 4.54468085106383e-06,
|
| 361 |
-
"loss": 0.0024,
|
| 362 |
-
"step": 29500
|
| 363 |
},
|
| 364 |
{
|
| 365 |
"epoch": 2.4,
|
| 366 |
-
"learning_rate":
|
| 367 |
-
"loss": 0.
|
| 368 |
-
"step":
|
| 369 |
-
},
|
| 370 |
-
{
|
| 371 |
-
"epoch": 2.44,
|
| 372 |
-
"learning_rate": 3.977872340425532e-06,
|
| 373 |
-
"loss": 0.0024,
|
| 374 |
-
"step": 30500
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 2.48,
|
| 378 |
-
"learning_rate":
|
| 379 |
-
"loss": 0.
|
| 380 |
-
"step":
|
| 381 |
-
},
|
| 382 |
-
{
|
| 383 |
-
"epoch": 2.52,
|
| 384 |
-
"learning_rate": 3.41049645390071e-06,
|
| 385 |
-
"loss": 0.0024,
|
| 386 |
-
"step": 31500
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 2.56,
|
| 390 |
-
"learning_rate":
|
| 391 |
-
"loss": 0.
|
| 392 |
-
"step":
|
| 393 |
-
},
|
| 394 |
-
{
|
| 395 |
-
"epoch": 2.6,
|
| 396 |
-
"learning_rate": 2.8436879432624116e-06,
|
| 397 |
-
"loss": 0.0024,
|
| 398 |
-
"step": 32500
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 2.64,
|
| 402 |
-
"learning_rate":
|
| 403 |
-
"loss": 0.
|
| 404 |
-
"step":
|
| 405 |
-
},
|
| 406 |
-
{
|
| 407 |
-
"epoch": 2.68,
|
| 408 |
-
"learning_rate": 2.276312056737589e-06,
|
| 409 |
-
"loss": 0.0024,
|
| 410 |
-
"step": 33500
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 2.72,
|
| 414 |
-
"learning_rate":
|
| 415 |
-
"loss": 0.
|
| 416 |
-
"step":
|
| 417 |
-
},
|
| 418 |
-
{
|
| 419 |
-
"epoch": 2.76,
|
| 420 |
-
"learning_rate": 1.709503546099291e-06,
|
| 421 |
-
"loss": 0.0024,
|
| 422 |
-
"step": 34500
|
| 423 |
},
|
| 424 |
{
|
| 425 |
"epoch": 2.8,
|
| 426 |
-
"learning_rate":
|
| 427 |
-
"loss": 0.
|
| 428 |
-
"step":
|
| 429 |
-
},
|
| 430 |
-
{
|
| 431 |
-
"epoch": 2.84,
|
| 432 |
-
"learning_rate": 1.1421276595744682e-06,
|
| 433 |
-
"loss": 0.0023,
|
| 434 |
-
"step": 35500
|
| 435 |
},
|
| 436 |
{
|
| 437 |
"epoch": 2.88,
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss": 0.
|
| 440 |
-
"step":
|
| 441 |
-
},
|
| 442 |
-
{
|
| 443 |
-
"epoch": 2.92,
|
| 444 |
-
"learning_rate": 5.753191489361703e-07,
|
| 445 |
-
"loss": 0.0023,
|
| 446 |
-
"step": 36500
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 2.96,
|
| 450 |
-
"learning_rate":
|
| 451 |
-
"loss": 0.
|
| 452 |
-
"step":
|
| 453 |
-
},
|
| 454 |
-
{
|
| 455 |
-
"epoch": 3.0,
|
| 456 |
-
"learning_rate": 7.943262411347518e-09,
|
| 457 |
-
"loss": 0.0024,
|
| 458 |
-
"step": 37500
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 3.0,
|
| 462 |
-
"step":
|
| 463 |
-
"total_flos":
|
| 464 |
-
"train_runtime":
|
| 465 |
-
"train_samples_per_second":
|
| 466 |
}
|
| 467 |
],
|
| 468 |
-
"max_steps":
|
| 469 |
"num_train_epochs": 3,
|
| 470 |
-
"total_flos":
|
| 471 |
"trial_name": null,
|
| 472 |
"trial_params": null
|
| 473 |
}
|
|
|
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 3.0,
|
| 5 |
+
"global_step": 18750,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
{
|
| 11 |
"epoch": 0.08,
|
| 12 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 13 |
+
"loss": 0.017,
|
| 14 |
+
"step": 500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.16,
|
| 18 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 19 |
+
"loss": 0.0087,
|
| 20 |
+
"step": 1000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.24,
|
| 24 |
+
"learning_rate": 4.893617021276596e-05,
|
| 25 |
+
"loss": 0.008,
|
| 26 |
+
"step": 1500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.32,
|
| 30 |
+
"learning_rate": 4.751773049645391e-05,
|
| 31 |
+
"loss": 0.008,
|
| 32 |
+
"step": 2000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.4,
|
| 36 |
+
"learning_rate": 4.609929078014185e-05,
|
| 37 |
+
"loss": 0.0082,
|
| 38 |
+
"step": 2500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.48,
|
| 42 |
+
"learning_rate": 4.468085106382979e-05,
|
| 43 |
+
"loss": 0.0077,
|
| 44 |
+
"step": 3000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.56,
|
| 48 |
+
"learning_rate": 4.3262411347517734e-05,
|
| 49 |
+
"loss": 0.0078,
|
| 50 |
+
"step": 3500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.64,
|
| 54 |
+
"learning_rate": 4.1843971631205674e-05,
|
| 55 |
+
"loss": 0.0077,
|
| 56 |
+
"step": 4000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 0.72,
|
| 60 |
+
"learning_rate": 4.0425531914893614e-05,
|
| 61 |
+
"loss": 0.0076,
|
| 62 |
+
"step": 4500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.8,
|
| 66 |
+
"learning_rate": 3.900709219858156e-05,
|
| 67 |
+
"loss": 0.0077,
|
| 68 |
+
"step": 5000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 0.88,
|
| 72 |
+
"learning_rate": 3.75886524822695e-05,
|
| 73 |
+
"loss": 0.0076,
|
| 74 |
+
"step": 5500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.96,
|
| 78 |
+
"learning_rate": 3.617021276595745e-05,
|
| 79 |
+
"loss": 0.0075,
|
| 80 |
+
"step": 6000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 1.04,
|
| 84 |
+
"learning_rate": 3.4754609929078014e-05,
|
| 85 |
+
"loss": 0.0074,
|
| 86 |
+
"step": 6500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 1.12,
|
| 90 |
+
"learning_rate": 3.333617021276596e-05,
|
| 91 |
+
"loss": 0.0069,
|
| 92 |
+
"step": 7000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 1.2,
|
| 96 |
+
"learning_rate": 3.19177304964539e-05,
|
| 97 |
+
"loss": 0.007,
|
| 98 |
+
"step": 7500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 1.28,
|
| 102 |
+
"learning_rate": 3.0499290780141848e-05,
|
| 103 |
+
"loss": 0.0069,
|
| 104 |
+
"step": 8000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 1.36,
|
| 108 |
+
"learning_rate": 2.908368794326241e-05,
|
| 109 |
+
"loss": 0.0068,
|
| 110 |
+
"step": 8500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 1.44,
|
| 114 |
+
"learning_rate": 2.7665248226950357e-05,
|
| 115 |
+
"loss": 0.0068,
|
| 116 |
+
"step": 9000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 1.52,
|
| 120 |
+
"learning_rate": 2.6246808510638298e-05,
|
| 121 |
+
"loss": 0.007,
|
| 122 |
+
"step": 9500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 1.6,
|
| 126 |
+
"learning_rate": 2.4828368794326244e-05,
|
| 127 |
+
"loss": 0.0069,
|
| 128 |
+
"step": 10000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 1.68,
|
| 132 |
+
"learning_rate": 2.3412765957446807e-05,
|
| 133 |
+
"loss": 0.0069,
|
| 134 |
+
"step": 10500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 135 |
},
|
| 136 |
{
|
| 137 |
"epoch": 1.76,
|
| 138 |
+
"learning_rate": 2.1994326241134754e-05,
|
| 139 |
+
"loss": 0.007,
|
| 140 |
+
"step": 11000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 141 |
},
|
| 142 |
{
|
| 143 |
"epoch": 1.84,
|
| 144 |
+
"learning_rate": 2.0575886524822697e-05,
|
| 145 |
+
"loss": 0.0068,
|
| 146 |
+
"step": 11500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 1.92,
|
| 150 |
+
"learning_rate": 1.915744680851064e-05,
|
| 151 |
+
"loss": 0.0069,
|
| 152 |
+
"step": 12000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 2.0,
|
| 156 |
+
"learning_rate": 1.773900709219858e-05,
|
| 157 |
+
"loss": 0.0068,
|
| 158 |
+
"step": 12500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 2.08,
|
| 162 |
+
"learning_rate": 1.6320567375886524e-05,
|
| 163 |
+
"loss": 0.0064,
|
| 164 |
+
"step": 13000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 2.16,
|
| 168 |
+
"learning_rate": 1.4904964539007094e-05,
|
| 169 |
+
"loss": 0.0064,
|
| 170 |
+
"step": 13500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 2.24,
|
| 174 |
+
"learning_rate": 1.3486524822695037e-05,
|
| 175 |
+
"loss": 0.0065,
|
| 176 |
+
"step": 14000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 2.32,
|
| 180 |
+
"learning_rate": 1.2068085106382979e-05,
|
| 181 |
+
"loss": 0.0064,
|
| 182 |
+
"step": 14500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 2.4,
|
| 186 |
+
"learning_rate": 1.0649645390070922e-05,
|
| 187 |
+
"loss": 0.0064,
|
| 188 |
+
"step": 15000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 2.48,
|
| 192 |
+
"learning_rate": 9.231205673758866e-06,
|
| 193 |
+
"loss": 0.0063,
|
| 194 |
+
"step": 15500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 2.56,
|
| 198 |
+
"learning_rate": 7.812765957446809e-06,
|
| 199 |
+
"loss": 0.0064,
|
| 200 |
+
"step": 16000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 2.64,
|
| 204 |
+
"learning_rate": 6.394326241134752e-06,
|
| 205 |
+
"loss": 0.0064,
|
| 206 |
+
"step": 16500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 2.72,
|
| 210 |
+
"learning_rate": 4.975886524822695e-06,
|
| 211 |
+
"loss": 0.0064,
|
| 212 |
+
"step": 17000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 2.8,
|
| 216 |
+
"learning_rate": 3.5602836879432625e-06,
|
| 217 |
+
"loss": 0.0063,
|
| 218 |
+
"step": 17500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 219 |
},
|
| 220 |
{
|
| 221 |
"epoch": 2.88,
|
| 222 |
+
"learning_rate": 2.1418439716312055e-06,
|
| 223 |
+
"loss": 0.0064,
|
| 224 |
+
"step": 18000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 225 |
},
|
| 226 |
{
|
| 227 |
"epoch": 2.96,
|
| 228 |
+
"learning_rate": 7.23404255319149e-07,
|
| 229 |
+
"loss": 0.0063,
|
| 230 |
+
"step": 18500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 3.0,
|
| 234 |
+
"step": 18750,
|
| 235 |
+
"total_flos": 0,
|
| 236 |
+
"train_runtime": 1200.1754,
|
| 237 |
+
"train_samples_per_second": 15.623
|
| 238 |
}
|
| 239 |
],
|
| 240 |
+
"max_steps": 18750,
|
| 241 |
"num_train_epochs": 3,
|
| 242 |
+
"total_flos": 0,
|
| 243 |
"trial_name": null,
|
| 244 |
"trial_params": null
|
| 245 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f86d7a02e6b1a4c491750e657c351e51deadb3844c96bca07e3808231dc2848c
|
| 3 |
+
size 2479
|