Upload folder using huggingface_hub
Browse files
12_mini_head_one_mask_per_head_1_latent_mask/log2.txt
CHANGED
|
@@ -2,528 +2,528 @@ max_steps: 4375
|
|
| 2 |
0 val loss 11.2876
|
| 3 |
0 val perplexity 79825.2969
|
| 4 |
0 train 11.284595 (lr=4.1958e-06) (hash(x)=93356070)
|
| 5 |
-
10 train 9.
|
| 6 |
-
20 train 9.
|
| 7 |
-
30 train 8.
|
| 8 |
-
40 train 7.
|
| 9 |
-
50 train 7.
|
| 10 |
-
60 train 7.
|
| 11 |
-
70 train 7.
|
| 12 |
-
80 train 7.
|
| 13 |
-
90 train 7.
|
| 14 |
-
100 val loss 7.
|
| 15 |
-
100 val perplexity
|
| 16 |
-
100 train 6.
|
| 17 |
-
110 train 6.
|
| 18 |
-
120 train 6.
|
| 19 |
-
130 train 6.
|
| 20 |
-
140 train 6.
|
| 21 |
-
150 train 6.
|
| 22 |
-
160 train 6.
|
| 23 |
-
170 train 6.
|
| 24 |
-
180 train 6.
|
| 25 |
-
190 train 6.
|
| 26 |
-
200 val loss 6.
|
| 27 |
-
200 val perplexity
|
| 28 |
-
200 train 6.
|
| 29 |
-
210 train 6.
|
| 30 |
-
220 train 6.
|
| 31 |
-
230 train 6.
|
| 32 |
-
240 train 5.
|
| 33 |
-
250 train 5.
|
| 34 |
-
260 train 5.
|
| 35 |
-
270 train 5.
|
| 36 |
-
280 train 5.
|
| 37 |
-
290 train 5.
|
| 38 |
-
300 val loss 5.
|
| 39 |
-
300 val perplexity
|
| 40 |
-
300 train 5.
|
| 41 |
-
310 train 5.
|
| 42 |
-
320 train 5.
|
| 43 |
-
330 train 5.
|
| 44 |
-
340 train 5.
|
| 45 |
-
350 train 5.
|
| 46 |
-
360 train 5.
|
| 47 |
-
370 train 5.
|
| 48 |
-
380 train 5.
|
| 49 |
-
390 train 5.
|
| 50 |
-
400 val loss 5.
|
| 51 |
-
400 val perplexity 229.
|
| 52 |
-
400 train 5.
|
| 53 |
-
410 train 5.
|
| 54 |
-
420 train 5.
|
| 55 |
-
430 train 5.
|
| 56 |
-
440 train 5.
|
| 57 |
-
450 train 5.
|
| 58 |
-
460 train 5.
|
| 59 |
-
470 train 4.
|
| 60 |
-
480 train 5.
|
| 61 |
-
490 train 4.
|
| 62 |
-
500 val loss 5.
|
| 63 |
-
500 val perplexity
|
| 64 |
-
500 train 5.
|
| 65 |
-
510 train 5.
|
| 66 |
-
520 train 5.
|
| 67 |
-
530 train 5.
|
| 68 |
-
540 train 5.
|
| 69 |
-
550 train 5.
|
| 70 |
-
560 train 5.
|
| 71 |
-
570 train 5.
|
| 72 |
-
580 train 5.
|
| 73 |
-
590 train 5.
|
| 74 |
-
600 val loss 5.
|
| 75 |
-
600 val perplexity
|
| 76 |
-
600 train 5.
|
| 77 |
-
610 train
|
| 78 |
-
620 train 4.
|
| 79 |
-
630 train 4.
|
| 80 |
-
640 train 4.
|
| 81 |
-
650 train 4.
|
| 82 |
-
660 train 4.
|
| 83 |
-
670 train 4.
|
| 84 |
-
680 train 4.
|
| 85 |
-
690 train 4.
|
| 86 |
-
700 val loss
|
| 87 |
-
700 val perplexity
|
| 88 |
-
700 train 4.
|
| 89 |
-
710 train 4.
|
| 90 |
-
720 train 4.
|
| 91 |
-
730 train 4.
|
| 92 |
-
740 train 4.
|
| 93 |
-
750 train 4.
|
| 94 |
-
760 train 4.
|
| 95 |
-
770 train 4.
|
| 96 |
-
780 train 4.
|
| 97 |
-
790 train
|
| 98 |
-
800 val loss 4.
|
| 99 |
-
800 val perplexity
|
| 100 |
-
800 train 4.
|
| 101 |
-
810 train 4.
|
| 102 |
-
820 train 4.
|
| 103 |
-
830 train 4.
|
| 104 |
-
840 train 4.
|
| 105 |
-
850 train 4.
|
| 106 |
-
860 train 4.
|
| 107 |
-
870 train
|
| 108 |
-
880 train
|
| 109 |
-
890 train
|
| 110 |
-
900 val loss
|
| 111 |
-
900 val perplexity
|
| 112 |
-
900 train
|
| 113 |
-
910 train
|
| 114 |
-
920 train
|
| 115 |
-
930 train
|
| 116 |
-
940 train
|
| 117 |
-
950 train 4.
|
| 118 |
-
960 train 4.
|
| 119 |
-
970 train 4.
|
| 120 |
-
980 train
|
| 121 |
-
990 train
|
| 122 |
-
1000 val loss
|
| 123 |
-
1000 val perplexity
|
| 124 |
-
1000 train
|
| 125 |
-
1010 train
|
| 126 |
-
1020 train
|
| 127 |
-
1030 train
|
| 128 |
-
1040 train
|
| 129 |
-
1050 train
|
| 130 |
-
1060 train
|
| 131 |
-
1070 train
|
| 132 |
-
1080 train
|
| 133 |
-
1090 train
|
| 134 |
-
1100 val loss
|
| 135 |
-
1100 val perplexity
|
| 136 |
-
1100 train
|
| 137 |
-
1110 train
|
| 138 |
-
1120 train
|
| 139 |
-
1130 train
|
| 140 |
-
1140 train
|
| 141 |
-
1150 train
|
| 142 |
-
1160 train
|
| 143 |
-
1170 train
|
| 144 |
-
1180 train
|
| 145 |
-
1190 train
|
| 146 |
-
1200 val loss
|
| 147 |
-
1200 val perplexity
|
| 148 |
-
1200 train
|
| 149 |
-
1210 train
|
| 150 |
-
1220 train
|
| 151 |
-
1230 train
|
| 152 |
-
1240 train
|
| 153 |
-
1250 train
|
| 154 |
-
1260 train
|
| 155 |
-
1270 train
|
| 156 |
-
1280 train
|
| 157 |
-
1290 train
|
| 158 |
-
1300 val loss
|
| 159 |
-
1300 val perplexity
|
| 160 |
-
1300 train
|
| 161 |
-
1310 train
|
| 162 |
-
1320 train
|
| 163 |
-
1330 train
|
| 164 |
-
1340 train
|
| 165 |
-
1350 train
|
| 166 |
-
1360 train
|
| 167 |
-
1370 train
|
| 168 |
-
1380 train
|
| 169 |
-
1390 train
|
| 170 |
-
1400 val loss
|
| 171 |
-
1400 val perplexity
|
| 172 |
-
1400 train
|
| 173 |
-
1410 train
|
| 174 |
-
1420 train
|
| 175 |
-
1430 train
|
| 176 |
-
1440 train
|
| 177 |
-
1450 train
|
| 178 |
-
1460 train
|
| 179 |
-
1470 train 5.
|
| 180 |
-
1480 train 5.
|
| 181 |
-
1490 train 5.
|
| 182 |
-
1500 val loss
|
| 183 |
-
1500 val perplexity
|
| 184 |
-
1500 train
|
| 185 |
-
1510 train
|
| 186 |
-
1520 train 5.
|
| 187 |
-
1530 train 5.
|
| 188 |
-
1540 train 5.
|
| 189 |
-
1550 train 5.
|
| 190 |
-
1560 train 5.
|
| 191 |
-
1570 train 5.
|
| 192 |
-
1580 train 5.
|
| 193 |
-
1590 train 5.
|
| 194 |
-
1600 val loss 5.
|
| 195 |
-
1600 val perplexity
|
| 196 |
-
1600 train 5.
|
| 197 |
-
1610 train 5.
|
| 198 |
-
1620 train 5.
|
| 199 |
-
1630 train 5.
|
| 200 |
-
1640 train 5.
|
| 201 |
-
1650 train 5.
|
| 202 |
-
1660 train 5.
|
| 203 |
-
1670 train 5.
|
| 204 |
-
1680 train 5.
|
| 205 |
-
1690 train 5.
|
| 206 |
-
1700 val loss 5.
|
| 207 |
-
1700 val perplexity
|
| 208 |
-
1700 train 5.
|
| 209 |
-
1710 train
|
| 210 |
-
1720 train 5.
|
| 211 |
-
1730 train
|
| 212 |
-
1740 train
|
| 213 |
-
1750 train
|
| 214 |
-
1760 train
|
| 215 |
-
1770 train 5.
|
| 216 |
-
1780 train
|
| 217 |
-
1790 train
|
| 218 |
-
1800 val loss
|
| 219 |
-
1800 val perplexity
|
| 220 |
-
1800 train
|
| 221 |
-
1810 train
|
| 222 |
-
1820 train
|
| 223 |
-
1830 train
|
| 224 |
-
1840 train
|
| 225 |
-
1850 train
|
| 226 |
-
1860 train
|
| 227 |
-
1870 train 6.
|
| 228 |
-
1880 train 6.
|
| 229 |
-
1890 train 6.
|
| 230 |
-
1900 val loss 6.
|
| 231 |
-
1900 val perplexity
|
| 232 |
-
1900 train 6.
|
| 233 |
-
1910 train 6.
|
| 234 |
-
1920 train 6.
|
| 235 |
-
1930 train 6.
|
| 236 |
-
1940 train 6.
|
| 237 |
-
1950 train 6.
|
| 238 |
-
1960 train 6.
|
| 239 |
-
1970 train
|
| 240 |
-
1980 train
|
| 241 |
-
1990 train
|
| 242 |
-
2000 val loss
|
| 243 |
-
2000 val perplexity
|
| 244 |
-
2000 train
|
| 245 |
-
2010 train
|
| 246 |
-
2020 train
|
| 247 |
-
2030 train
|
| 248 |
-
2040 train
|
| 249 |
-
2050 train
|
| 250 |
-
2060 train 6.
|
| 251 |
-
2070 train
|
| 252 |
-
2080 train
|
| 253 |
-
2090 train 6.
|
| 254 |
-
2100 val loss 6.
|
| 255 |
-
2100 val perplexity
|
| 256 |
-
2100 train 6.
|
| 257 |
-
2110 train
|
| 258 |
-
2120 train 6.
|
| 259 |
-
2130 train 6.
|
| 260 |
-
2140 train
|
| 261 |
-
2150 train 6.
|
| 262 |
-
2160 train 6.
|
| 263 |
-
2170 train 6.
|
| 264 |
-
2180 train
|
| 265 |
-
2190 train 6.
|
| 266 |
-
2200 val loss 6.
|
| 267 |
-
2200 val perplexity
|
| 268 |
-
2200 train 6.
|
| 269 |
-
2210 train 6.
|
| 270 |
-
2220 train 6.
|
| 271 |
-
2230 train
|
| 272 |
-
2240 train 6.
|
| 273 |
-
2250 train 6.
|
| 274 |
-
2260 train 6.
|
| 275 |
-
2270 train 6.
|
| 276 |
-
2280 train 6.
|
| 277 |
-
2290 train 6.
|
| 278 |
-
2300 val loss 6.
|
| 279 |
-
2300 val perplexity
|
| 280 |
-
2300 train 6.
|
| 281 |
-
2310 train
|
| 282 |
-
2320 train
|
| 283 |
-
2330 train
|
| 284 |
-
2340 train
|
| 285 |
-
2350 train
|
| 286 |
-
2360 train
|
| 287 |
-
2370 train
|
| 288 |
-
2380 train
|
| 289 |
-
2390 train
|
| 290 |
-
2400 val loss
|
| 291 |
-
2400 val perplexity
|
| 292 |
-
2400 train
|
| 293 |
-
2410 train
|
| 294 |
-
2420 train
|
| 295 |
-
2430 train
|
| 296 |
-
2440 train
|
| 297 |
-
2450 train
|
| 298 |
-
2460 train
|
| 299 |
-
2470 train
|
| 300 |
-
2480 train
|
| 301 |
-
2490 train
|
| 302 |
-
2500 val loss
|
| 303 |
-
2500 val perplexity
|
| 304 |
-
2500 train
|
| 305 |
-
2510 train
|
| 306 |
-
2520 train
|
| 307 |
-
2530 train
|
| 308 |
-
2540 train
|
| 309 |
-
2550 train
|
| 310 |
-
2560 train
|
| 311 |
-
2570 train
|
| 312 |
-
2580 train
|
| 313 |
-
2590 train
|
| 314 |
-
2600 val loss
|
| 315 |
-
2600 val perplexity
|
| 316 |
-
2600 train
|
| 317 |
-
2610 train
|
| 318 |
-
2620 train
|
| 319 |
-
2630 train
|
| 320 |
-
2640 train
|
| 321 |
-
2650 train
|
| 322 |
-
2660 train
|
| 323 |
-
2670 train 7.
|
| 324 |
-
2680 train 7.
|
| 325 |
-
2690 train 7.
|
| 326 |
-
2700 val loss
|
| 327 |
-
2700 val perplexity
|
| 328 |
-
2700 train
|
| 329 |
-
2710 train
|
| 330 |
-
2720 train
|
| 331 |
-
2730 train
|
| 332 |
-
2740 train
|
| 333 |
-
2750 train
|
| 334 |
-
2760 train
|
| 335 |
-
2770 train
|
| 336 |
-
2780 train
|
| 337 |
-
2790 train
|
| 338 |
-
2800 val loss
|
| 339 |
-
2800 val perplexity
|
| 340 |
-
2800 train
|
| 341 |
-
2810 train 7.
|
| 342 |
-
2820 train 7.
|
| 343 |
-
2830 train
|
| 344 |
-
2840 train
|
| 345 |
-
2850 train 7.
|
| 346 |
-
2860 train 7.
|
| 347 |
-
2870 train
|
| 348 |
-
2880 train
|
| 349 |
-
2890 train
|
| 350 |
-
2900 val loss 7.
|
| 351 |
-
2900 val perplexity
|
| 352 |
-
2900 train
|
| 353 |
-
2910 train
|
| 354 |
-
2920 train
|
| 355 |
-
2930 train 7.
|
| 356 |
-
2940 train
|
| 357 |
-
2950 train 7.
|
| 358 |
-
2960 train
|
| 359 |
-
2970 train
|
| 360 |
-
2980 train
|
| 361 |
-
2990 train
|
| 362 |
-
3000 val loss
|
| 363 |
-
3000 val perplexity
|
| 364 |
-
3000 train
|
| 365 |
-
3010 train
|
| 366 |
-
3020 train
|
| 367 |
-
3030 train
|
| 368 |
-
3040 train
|
| 369 |
-
3050 train
|
| 370 |
-
3060 train
|
| 371 |
-
3070 train
|
| 372 |
-
3080 train
|
| 373 |
-
3090 train
|
| 374 |
-
3100 val loss
|
| 375 |
-
3100 val perplexity
|
| 376 |
-
3100 train
|
| 377 |
-
3110 train
|
| 378 |
-
3120 train
|
| 379 |
-
3130 train
|
| 380 |
-
3140 train
|
| 381 |
-
3150 train
|
| 382 |
-
3160 train
|
| 383 |
-
3170 train 7.
|
| 384 |
-
3180 train 7.
|
| 385 |
-
3190 train
|
| 386 |
-
3200 val loss 7.
|
| 387 |
-
3200 val perplexity
|
| 388 |
-
3200 train 7.
|
| 389 |
-
3210 train 7.
|
| 390 |
-
3220 train
|
| 391 |
-
3230 train
|
| 392 |
-
3240 train
|
| 393 |
-
3250 train
|
| 394 |
-
3260 train
|
| 395 |
-
3270 train
|
| 396 |
-
3280 train
|
| 397 |
-
3290 train
|
| 398 |
-
3300 val loss
|
| 399 |
-
3300 val perplexity
|
| 400 |
-
3300 train
|
| 401 |
-
3310 train
|
| 402 |
-
3320 train
|
| 403 |
-
3330 train
|
| 404 |
-
3340 train
|
| 405 |
-
3350 train
|
| 406 |
-
3360 train
|
| 407 |
-
3370 train
|
| 408 |
-
3380 train
|
| 409 |
-
3390 train 7.
|
| 410 |
-
3400 val loss
|
| 411 |
-
3400 val perplexity
|
| 412 |
-
3400 train
|
| 413 |
-
3410 train
|
| 414 |
-
3420 train
|
| 415 |
-
3430 train
|
| 416 |
-
3440 train
|
| 417 |
-
3450 train
|
| 418 |
-
3460 train
|
| 419 |
-
3470 train
|
| 420 |
-
3480 train
|
| 421 |
-
3490 train
|
| 422 |
-
3500 val loss
|
| 423 |
-
3500 val perplexity
|
| 424 |
-
3500 train
|
| 425 |
-
3510 train
|
| 426 |
-
3520 train
|
| 427 |
-
3530 train
|
| 428 |
-
3540 train 7.
|
| 429 |
-
3550 train
|
| 430 |
-
3560 train
|
| 431 |
-
3570 train
|
| 432 |
-
3580 train
|
| 433 |
-
3590 train
|
| 434 |
-
3600 val loss
|
| 435 |
-
3600 val perplexity
|
| 436 |
-
3600 train
|
| 437 |
-
3610 train
|
| 438 |
-
3620 train
|
| 439 |
-
3630 train
|
| 440 |
-
3640 train
|
| 441 |
-
3650 train
|
| 442 |
-
3660 train
|
| 443 |
-
3670 train
|
| 444 |
-
3680 train
|
| 445 |
-
3690 train 7.
|
| 446 |
-
3700 val loss 7.
|
| 447 |
-
3700 val perplexity
|
| 448 |
-
3700 train
|
| 449 |
-
3710 train 7.
|
| 450 |
-
3720 train
|
| 451 |
-
3730 train
|
| 452 |
-
3740 train
|
| 453 |
-
3750 train
|
| 454 |
-
3760 train
|
| 455 |
-
3770 train
|
| 456 |
-
3780 train 7.
|
| 457 |
-
3790 train
|
| 458 |
-
3800 val loss 7.
|
| 459 |
-
3800 val perplexity
|
| 460 |
-
3800 train
|
| 461 |
-
3810 train
|
| 462 |
-
3820 train
|
| 463 |
-
3830 train
|
| 464 |
-
3840 train
|
| 465 |
-
3850 train
|
| 466 |
-
3860 train
|
| 467 |
-
3870 train
|
| 468 |
-
3880 train
|
| 469 |
-
3890 train 7.
|
| 470 |
-
3900 val loss
|
| 471 |
-
3900 val perplexity
|
| 472 |
-
3900 train
|
| 473 |
-
3910 train
|
| 474 |
-
3920 train
|
| 475 |
-
3930 train
|
| 476 |
-
3940 train
|
| 477 |
-
3950 train
|
| 478 |
-
3960 train
|
| 479 |
-
3970 train
|
| 480 |
-
3980 train
|
| 481 |
-
3990 train
|
| 482 |
-
4000 val loss
|
| 483 |
-
4000 val perplexity
|
| 484 |
-
4000 train
|
| 485 |
-
4010 train
|
| 486 |
-
4020 train
|
| 487 |
-
4030 train
|
| 488 |
-
4040 train
|
| 489 |
-
4050 train
|
| 490 |
-
4060 train
|
| 491 |
-
4070 train
|
| 492 |
-
4080 train
|
| 493 |
-
4090 train
|
| 494 |
-
4100 val loss
|
| 495 |
-
4100 val perplexity
|
| 496 |
-
4100 train
|
| 497 |
-
4110 train
|
| 498 |
-
4120 train
|
| 499 |
-
4130 train
|
| 500 |
-
4140 train
|
| 501 |
-
4150 train
|
| 502 |
-
4160 train
|
| 503 |
-
4170 train
|
| 504 |
-
4180 train
|
| 505 |
-
4190 train
|
| 506 |
-
4200 val loss
|
| 507 |
-
4200 val perplexity
|
| 508 |
-
4200 train
|
| 509 |
-
4210 train
|
| 510 |
-
4220 train
|
| 511 |
-
4230 train
|
| 512 |
-
4240 train
|
| 513 |
-
4250 train
|
| 514 |
-
4260 train
|
| 515 |
-
4270 train
|
| 516 |
-
4280 train
|
| 517 |
-
4290 train
|
| 518 |
-
4300 val loss
|
| 519 |
-
4300 val perplexity
|
| 520 |
-
4300 train
|
| 521 |
-
4310 train
|
| 522 |
-
4320 train
|
| 523 |
-
4330 train
|
| 524 |
-
4340 train
|
| 525 |
-
4350 train
|
| 526 |
-
4360 train
|
| 527 |
-
4370 train
|
| 528 |
-
4374 val loss
|
| 529 |
-
4374 val perplexity
|
|
|
|
| 2 |
0 val loss 11.2876
|
| 3 |
0 val perplexity 79825.2969
|
| 4 |
0 train 11.284595 (lr=4.1958e-06) (hash(x)=93356070)
|
| 5 |
+
10 train 9.923347 (lr=4.6154e-05) (hash(x)=91253010)
|
| 6 |
+
20 train 9.354902 (lr=8.8112e-05) (hash(x)=74934453)
|
| 7 |
+
30 train 8.647146 (lr=1.3007e-04) (hash(x)=79278034)
|
| 8 |
+
40 train 7.918366 (lr=1.7203e-04) (hash(x)=80107892)
|
| 9 |
+
50 train 7.701303 (lr=2.1399e-04) (hash(x)=70745428)
|
| 10 |
+
60 train 7.750876 (lr=2.5594e-04) (hash(x)=80077589)
|
| 11 |
+
70 train 7.409539 (lr=2.9790e-04) (hash(x)=76213766)
|
| 12 |
+
80 train 7.397988 (lr=3.3986e-04) (hash(x)=83218328)
|
| 13 |
+
90 train 7.214713 (lr=3.8182e-04) (hash(x)=74855845)
|
| 14 |
+
100 val loss 7.1211
|
| 15 |
+
100 val perplexity 1237.8484
|
| 16 |
+
100 train 6.954023 (lr=4.2378e-04) (hash(x)=82814902)
|
| 17 |
+
110 train 6.924104 (lr=4.6573e-04) (hash(x)=84907741)
|
| 18 |
+
120 train 6.753369 (lr=5.0769e-04) (hash(x)=82613223)
|
| 19 |
+
130 train 6.601605 (lr=5.4965e-04) (hash(x)=83540876)
|
| 20 |
+
140 train 6.589177 (lr=5.9161e-04) (hash(x)=75095216)
|
| 21 |
+
150 train 6.661979 (lr=6.3357e-04) (hash(x)=97190944)
|
| 22 |
+
160 train 6.576634 (lr=6.7552e-04) (hash(x)=82117809)
|
| 23 |
+
170 train 6.490225 (lr=7.1748e-04) (hash(x)=70514724)
|
| 24 |
+
180 train 6.416883 (lr=7.5944e-04) (hash(x)=81029624)
|
| 25 |
+
190 train 6.324186 (lr=8.0140e-04) (hash(x)=78996841)
|
| 26 |
+
200 val loss 6.2581
|
| 27 |
+
200 val perplexity 522.2218
|
| 28 |
+
200 train 6.260619 (lr=8.4336e-04) (hash(x)=79845097)
|
| 29 |
+
210 train 6.152258 (lr=8.8531e-04) (hash(x)=78997683)
|
| 30 |
+
220 train 6.074738 (lr=9.2727e-04) (hash(x)=74895865)
|
| 31 |
+
230 train 6.206913 (lr=9.6923e-04) (hash(x)=80933276)
|
| 32 |
+
240 train 5.901365 (lr=1.0112e-03) (hash(x)=77664606)
|
| 33 |
+
250 train 5.960237 (lr=1.0531e-03) (hash(x)=75903930)
|
| 34 |
+
260 train 5.842309 (lr=1.0951e-03) (hash(x)=82996853)
|
| 35 |
+
270 train 5.776155 (lr=1.1371e-03) (hash(x)=73269568)
|
| 36 |
+
280 train 5.717534 (lr=1.1790e-03) (hash(x)=82906427)
|
| 37 |
+
290 train 5.636944 (lr=1.2210e-03) (hash(x)=71797895)
|
| 38 |
+
300 val loss 5.7906
|
| 39 |
+
300 val perplexity 327.2250
|
| 40 |
+
300 train 5.498541 (lr=1.2629e-03) (hash(x)=85232249)
|
| 41 |
+
310 train 5.447090 (lr=1.3049e-03) (hash(x)=75252489)
|
| 42 |
+
320 train 5.526665 (lr=1.3469e-03) (hash(x)=74394644)
|
| 43 |
+
330 train 5.625390 (lr=1.3888e-03) (hash(x)=82772910)
|
| 44 |
+
340 train 5.669609 (lr=1.4308e-03) (hash(x)=81627464)
|
| 45 |
+
350 train 5.644387 (lr=1.4727e-03) (hash(x)=84678053)
|
| 46 |
+
360 train 5.578918 (lr=1.5147e-03) (hash(x)=81884128)
|
| 47 |
+
370 train 5.590245 (lr=1.5566e-03) (hash(x)=76379242)
|
| 48 |
+
380 train 5.401692 (lr=1.5986e-03) (hash(x)=98182875)
|
| 49 |
+
390 train 5.377892 (lr=1.6406e-03) (hash(x)=79710436)
|
| 50 |
+
400 val loss 5.4353
|
| 51 |
+
400 val perplexity 229.3728
|
| 52 |
+
400 train 5.375259 (lr=1.6825e-03) (hash(x)=79841071)
|
| 53 |
+
410 train 5.329089 (lr=1.7245e-03) (hash(x)=75844151)
|
| 54 |
+
420 train 5.335932 (lr=1.7664e-03) (hash(x)=73125036)
|
| 55 |
+
430 train 5.366583 (lr=1.8084e-03) (hash(x)=84214858)
|
| 56 |
+
440 train 5.262486 (lr=1.8503e-03) (hash(x)=80456994)
|
| 57 |
+
450 train 5.248571 (lr=1.8923e-03) (hash(x)=76735962)
|
| 58 |
+
460 train 5.157775 (lr=1.9343e-03) (hash(x)=81845446)
|
| 59 |
+
470 train 4.963287 (lr=1.9762e-03) (hash(x)=76094689)
|
| 60 |
+
480 train 5.126496 (lr=2.0182e-03) (hash(x)=83806686)
|
| 61 |
+
490 train 4.912891 (lr=2.0601e-03) (hash(x)=84690227)
|
| 62 |
+
500 val loss 5.2835
|
| 63 |
+
500 val perplexity 197.0522
|
| 64 |
+
500 train 5.015106 (lr=2.1021e-03) (hash(x)=71851938)
|
| 65 |
+
510 train 5.131371 (lr=2.1441e-03) (hash(x)=77159346)
|
| 66 |
+
520 train 5.166546 (lr=2.1860e-03) (hash(x)=80755753)
|
| 67 |
+
530 train 5.287557 (lr=2.2280e-03) (hash(x)=75379679)
|
| 68 |
+
540 train 5.182461 (lr=2.2699e-03) (hash(x)=82458619)
|
| 69 |
+
550 train 5.139487 (lr=2.3119e-03) (hash(x)=86366396)
|
| 70 |
+
560 train 5.191271 (lr=2.3538e-03) (hash(x)=85095044)
|
| 71 |
+
570 train 5.153404 (lr=2.3958e-03) (hash(x)=78385159)
|
| 72 |
+
580 train 5.083260 (lr=2.4378e-03) (hash(x)=79342394)
|
| 73 |
+
590 train 5.093650 (lr=2.4797e-03) (hash(x)=70782192)
|
| 74 |
+
600 val loss 5.1002
|
| 75 |
+
600 val perplexity 164.0547
|
| 76 |
+
600 train 5.074724 (lr=2.5217e-03) (hash(x)=76130353)
|
| 77 |
+
610 train 4.996155 (lr=2.5636e-03) (hash(x)=74778440)
|
| 78 |
+
620 train 4.971506 (lr=2.6056e-03) (hash(x)=79129709)
|
| 79 |
+
630 train 4.871780 (lr=2.6476e-03) (hash(x)=76469962)
|
| 80 |
+
640 train 4.855979 (lr=2.6895e-03) (hash(x)=78288049)
|
| 81 |
+
650 train 4.814961 (lr=2.7315e-03) (hash(x)=76641388)
|
| 82 |
+
660 train 4.728936 (lr=2.7734e-03) (hash(x)=75937906)
|
| 83 |
+
670 train 4.753763 (lr=2.8154e-03) (hash(x)=74807157)
|
| 84 |
+
680 train 4.650416 (lr=2.8573e-03) (hash(x)=77490144)
|
| 85 |
+
690 train 4.670358 (lr=2.8993e-03) (hash(x)=72900124)
|
| 86 |
+
700 val loss 4.9652
|
| 87 |
+
700 val perplexity 143.3429
|
| 88 |
+
700 train 4.646469 (lr=2.9413e-03) (hash(x)=73424218)
|
| 89 |
+
710 train 4.929185 (lr=2.9832e-03) (hash(x)=86052345)
|
| 90 |
+
720 train 4.906153 (lr=3.0000e-03) (hash(x)=77462613)
|
| 91 |
+
730 train 4.901950 (lr=2.9999e-03) (hash(x)=79027471)
|
| 92 |
+
740 train 4.863056 (lr=2.9997e-03) (hash(x)=78149992)
|
| 93 |
+
750 train 4.811694 (lr=2.9994e-03) (hash(x)=89147499)
|
| 94 |
+
760 train 4.852089 (lr=2.9990e-03) (hash(x)=74931538)
|
| 95 |
+
770 train 4.797495 (lr=2.9985e-03) (hash(x)=89648204)
|
| 96 |
+
780 train 4.841113 (lr=2.9979e-03) (hash(x)=83708293)
|
| 97 |
+
790 train 4.863065 (lr=2.9972e-03) (hash(x)=84336512)
|
| 98 |
+
800 val loss 4.8232
|
| 99 |
+
800 val perplexity 124.3641
|
| 100 |
+
800 train 4.743608 (lr=2.9964e-03) (hash(x)=75025285)
|
| 101 |
+
810 train 4.655568 (lr=2.9955e-03) (hash(x)=79436536)
|
| 102 |
+
820 train 4.662436 (lr=2.9945e-03) (hash(x)=90467390)
|
| 103 |
+
830 train 4.702021 (lr=2.9934e-03) (hash(x)=77292020)
|
| 104 |
+
840 train 4.634523 (lr=2.9922e-03) (hash(x)=75568927)
|
| 105 |
+
850 train 4.663900 (lr=2.9909e-03) (hash(x)=79671781)
|
| 106 |
+
860 train 4.765933 (lr=2.9896e-03) (hash(x)=84280943)
|
| 107 |
+
870 train 4.733632 (lr=2.9881e-03) (hash(x)=80901294)
|
| 108 |
+
880 train 4.868389 (lr=2.9865e-03) (hash(x)=81437584)
|
| 109 |
+
890 train 4.710431 (lr=2.9848e-03) (hash(x)=73893778)
|
| 110 |
+
900 val loss 4.7302
|
| 111 |
+
900 val perplexity 113.3208
|
| 112 |
+
900 train 4.665327 (lr=2.9830e-03) (hash(x)=74987794)
|
| 113 |
+
910 train 4.678456 (lr=2.9811e-03) (hash(x)=93721374)
|
| 114 |
+
920 train 4.670795 (lr=2.9792e-03) (hash(x)=79149678)
|
| 115 |
+
930 train 4.630743 (lr=2.9771e-03) (hash(x)=83179387)
|
| 116 |
+
940 train 4.629661 (lr=2.9749e-03) (hash(x)=89009978)
|
| 117 |
+
950 train 4.612129 (lr=2.9726e-03) (hash(x)=76483735)
|
| 118 |
+
960 train 4.558267 (lr=2.9703e-03) (hash(x)=78638850)
|
| 119 |
+
970 train 4.591957 (lr=2.9678e-03) (hash(x)=90415206)
|
| 120 |
+
980 train 4.529873 (lr=2.9652e-03) (hash(x)=90819195)
|
| 121 |
+
990 train 4.651834 (lr=2.9626e-03) (hash(x)=85938852)
|
| 122 |
+
1000 val loss 4.6510
|
| 123 |
+
1000 val perplexity 104.6945
|
| 124 |
+
1000 train 4.700669 (lr=2.9598e-03) (hash(x)=82436789)
|
| 125 |
+
1010 train 4.752086 (lr=2.9570e-03) (hash(x)=63022149)
|
| 126 |
+
1020 train 4.668364 (lr=2.9540e-03) (hash(x)=83501199)
|
| 127 |
+
1030 train 4.655666 (lr=2.9510e-03) (hash(x)=77093326)
|
| 128 |
+
1040 train 4.542048 (lr=2.9478e-03) (hash(x)=73437559)
|
| 129 |
+
1050 train 4.633110 (lr=2.9446e-03) (hash(x)=84550388)
|
| 130 |
+
1060 train 4.637328 (lr=2.9412e-03) (hash(x)=86184566)
|
| 131 |
+
1070 train 4.626502 (lr=2.9378e-03) (hash(x)=85644922)
|
| 132 |
+
1080 train 4.574726 (lr=2.9343e-03) (hash(x)=86179801)
|
| 133 |
+
1090 train 4.566450 (lr=2.9307e-03) (hash(x)=82868303)
|
| 134 |
+
1100 val loss 4.6538
|
| 135 |
+
1100 val perplexity 104.9812
|
| 136 |
+
1100 train 4.672271 (lr=2.9270e-03) (hash(x)=96780388)
|
| 137 |
+
1110 train 4.617107 (lr=2.9231e-03) (hash(x)=87223122)
|
| 138 |
+
1120 train 4.604944 (lr=2.9192e-03) (hash(x)=77292786)
|
| 139 |
+
1130 train 4.702131 (lr=2.9152e-03) (hash(x)=88761375)
|
| 140 |
+
1140 train 4.580485 (lr=2.9112e-03) (hash(x)=80956468)
|
| 141 |
+
1150 train 4.735983 (lr=2.9070e-03) (hash(x)=84725462)
|
| 142 |
+
1160 train 4.745145 (lr=2.9027e-03) (hash(x)=76098113)
|
| 143 |
+
1170 train 4.863050 (lr=2.8983e-03) (hash(x)=84387685)
|
| 144 |
+
1180 train 4.763179 (lr=2.8939e-03) (hash(x)=76120321)
|
| 145 |
+
1190 train 4.727228 (lr=2.8893e-03) (hash(x)=85974065)
|
| 146 |
+
1200 val loss 4.7403
|
| 147 |
+
1200 val perplexity 114.4647
|
| 148 |
+
1200 train 4.617321 (lr=2.8847e-03) (hash(x)=79340644)
|
| 149 |
+
1210 train 4.740779 (lr=2.8800e-03) (hash(x)=88345025)
|
| 150 |
+
1220 train 4.660150 (lr=2.8751e-03) (hash(x)=86406230)
|
| 151 |
+
1230 train 4.674973 (lr=2.8702e-03) (hash(x)=80608975)
|
| 152 |
+
1240 train 4.730783 (lr=2.8652e-03) (hash(x)=81635225)
|
| 153 |
+
1250 train 4.606773 (lr=2.8601e-03) (hash(x)=82126376)
|
| 154 |
+
1260 train 4.847933 (lr=2.8550e-03) (hash(x)=83737972)
|
| 155 |
+
1270 train 4.736798 (lr=2.8497e-03) (hash(x)=81132345)
|
| 156 |
+
1280 train 4.831640 (lr=2.8443e-03) (hash(x)=85850404)
|
| 157 |
+
1290 train 4.842688 (lr=2.8389e-03) (hash(x)=83490640)
|
| 158 |
+
1300 val loss 4.8869
|
| 159 |
+
1300 val perplexity 132.5398
|
| 160 |
+
1300 train 4.850279 (lr=2.8333e-03) (hash(x)=77545187)
|
| 161 |
+
1310 train 4.926689 (lr=2.8277e-03) (hash(x)=86412685)
|
| 162 |
+
1320 train 4.820294 (lr=2.8220e-03) (hash(x)=86429640)
|
| 163 |
+
1330 train 5.377491 (lr=2.8162e-03) (hash(x)=79645281)
|
| 164 |
+
1340 train 5.139881 (lr=2.8103e-03) (hash(x)=82668541)
|
| 165 |
+
1350 train 4.888584 (lr=2.8044e-03) (hash(x)=73627248)
|
| 166 |
+
1360 train 4.907205 (lr=2.7983e-03) (hash(x)=74845133)
|
| 167 |
+
1370 train 4.846905 (lr=2.7922e-03) (hash(x)=77850497)
|
| 168 |
+
1380 train 4.907204 (lr=2.7860e-03) (hash(x)=76143954)
|
| 169 |
+
1390 train 4.939134 (lr=2.7797e-03) (hash(x)=80675544)
|
| 170 |
+
1400 val loss 5.0477
|
| 171 |
+
1400 val perplexity 155.6612
|
| 172 |
+
1400 train 4.953867 (lr=2.7733e-03) (hash(x)=76010938)
|
| 173 |
+
1410 train 5.045611 (lr=2.7668e-03) (hash(x)=81260300)
|
| 174 |
+
1420 train 5.154567 (lr=2.7603e-03) (hash(x)=83803629)
|
| 175 |
+
1430 train 5.031479 (lr=2.7536e-03) (hash(x)=83840110)
|
| 176 |
+
1440 train 5.303030 (lr=2.7469e-03) (hash(x)=97711831)
|
| 177 |
+
1450 train 5.274389 (lr=2.7401e-03) (hash(x)=81110139)
|
| 178 |
+
1460 train 5.209787 (lr=2.7333e-03) (hash(x)=90942839)
|
| 179 |
+
1470 train 5.231742 (lr=2.7263e-03) (hash(x)=80721564)
|
| 180 |
+
1480 train 5.429221 (lr=2.7193e-03) (hash(x)=77852059)
|
| 181 |
+
1490 train 5.408493 (lr=2.7121e-03) (hash(x)=80011365)
|
| 182 |
+
1500 val loss 5.4280
|
| 183 |
+
1500 val perplexity 227.6901
|
| 184 |
+
1500 train 5.405737 (lr=2.7050e-03) (hash(x)=84496142)
|
| 185 |
+
1510 train 5.299591 (lr=2.6977e-03) (hash(x)=70186729)
|
| 186 |
+
1520 train 5.411904 (lr=2.6903e-03) (hash(x)=74854227)
|
| 187 |
+
1530 train 5.428205 (lr=2.6829e-03) (hash(x)=77468161)
|
| 188 |
+
1540 train 5.487059 (lr=2.6754e-03) (hash(x)=87853059)
|
| 189 |
+
1550 train 5.384209 (lr=2.6678e-03) (hash(x)=71225436)
|
| 190 |
+
1560 train 5.382601 (lr=2.6602e-03) (hash(x)=84072783)
|
| 191 |
+
1570 train 5.544269 (lr=2.6525e-03) (hash(x)=72723098)
|
| 192 |
+
1580 train 5.437517 (lr=2.6447e-03) (hash(x)=90409866)
|
| 193 |
+
1590 train 5.584054 (lr=2.6368e-03) (hash(x)=87481378)
|
| 194 |
+
1600 val loss 5.4516
|
| 195 |
+
1600 val perplexity 233.1381
|
| 196 |
+
1600 train 5.482508 (lr=2.6289e-03) (hash(x)=77643862)
|
| 197 |
+
1610 train 5.410114 (lr=2.6208e-03) (hash(x)=88862575)
|
| 198 |
+
1620 train 5.318450 (lr=2.6128e-03) (hash(x)=84612581)
|
| 199 |
+
1630 train 5.356172 (lr=2.6046e-03) (hash(x)=87075989)
|
| 200 |
+
1640 train 5.224185 (lr=2.5964e-03) (hash(x)=88277361)
|
| 201 |
+
1650 train 5.325086 (lr=2.5881e-03) (hash(x)=78750236)
|
| 202 |
+
1660 train 5.268503 (lr=2.5797e-03) (hash(x)=82604581)
|
| 203 |
+
1670 train 5.132612 (lr=2.5713e-03) (hash(x)=68482265)
|
| 204 |
+
1680 train 5.205066 (lr=2.5628e-03) (hash(x)=75088835)
|
| 205 |
+
1690 train 5.178228 (lr=2.5542e-03) (hash(x)=66060989)
|
| 206 |
+
1700 val loss 5.5331
|
| 207 |
+
1700 val perplexity 252.9160
|
| 208 |
+
1700 train 5.492572 (lr=2.5455e-03) (hash(x)=79986754)
|
| 209 |
+
1710 train 5.452727 (lr=2.5368e-03) (hash(x)=83657930)
|
| 210 |
+
1720 train 5.499579 (lr=2.5281e-03) (hash(x)=81754135)
|
| 211 |
+
1730 train 5.419930 (lr=2.5192e-03) (hash(x)=78904427)
|
| 212 |
+
1740 train 5.450576 (lr=2.5103e-03) (hash(x)=85920177)
|
| 213 |
+
1750 train 5.402933 (lr=2.5014e-03) (hash(x)=86573211)
|
| 214 |
+
1760 train 5.331089 (lr=2.4924e-03) (hash(x)=81737128)
|
| 215 |
+
1770 train 5.446370 (lr=2.4833e-03) (hash(x)=77400968)
|
| 216 |
+
1780 train 5.519640 (lr=2.4741e-03) (hash(x)=73545497)
|
| 217 |
+
1790 train 5.650542 (lr=2.4649e-03) (hash(x)=71641943)
|
| 218 |
+
1800 val loss 5.6234
|
| 219 |
+
1800 val perplexity 276.8281
|
| 220 |
+
1800 train 5.624428 (lr=2.4556e-03) (hash(x)=87819781)
|
| 221 |
+
1810 train 5.547323 (lr=2.4463e-03) (hash(x)=86870770)
|
| 222 |
+
1820 train 5.558342 (lr=2.4369e-03) (hash(x)=82522211)
|
| 223 |
+
1830 train 5.575167 (lr=2.4275e-03) (hash(x)=61947437)
|
| 224 |
+
1840 train 5.627659 (lr=2.4180e-03) (hash(x)=79865406)
|
| 225 |
+
1850 train 5.870785 (lr=2.4084e-03) (hash(x)=79828721)
|
| 226 |
+
1860 train 5.923368 (lr=2.3988e-03) (hash(x)=80869571)
|
| 227 |
+
1870 train 6.123144 (lr=2.3891e-03) (hash(x)=73780971)
|
| 228 |
+
1880 train 6.105158 (lr=2.3794e-03) (hash(x)=79249549)
|
| 229 |
+
1890 train 6.573531 (lr=2.3696e-03) (hash(x)=81041904)
|
| 230 |
+
1900 val loss 6.5748
|
| 231 |
+
1900 val perplexity 716.7759
|
| 232 |
+
1900 train 6.612405 (lr=2.3598e-03) (hash(x)=82456430)
|
| 233 |
+
1910 train 6.260543 (lr=2.3499e-03) (hash(x)=82222135)
|
| 234 |
+
1920 train 6.345827 (lr=2.3400e-03) (hash(x)=70033249)
|
| 235 |
+
1930 train 6.255967 (lr=2.3300e-03) (hash(x)=72887360)
|
| 236 |
+
1940 train 6.373764 (lr=2.3200e-03) (hash(x)=83251100)
|
| 237 |
+
1950 train 6.557952 (lr=2.3099e-03) (hash(x)=79660266)
|
| 238 |
+
1960 train 6.232997 (lr=2.2998e-03) (hash(x)=78919068)
|
| 239 |
+
1970 train 6.191888 (lr=2.2896e-03) (hash(x)=94690431)
|
| 240 |
+
1980 train 6.047931 (lr=2.2793e-03) (hash(x)=88153756)
|
| 241 |
+
1990 train 6.408006 (lr=2.2691e-03) (hash(x)=77195688)
|
| 242 |
+
2000 val loss 6.2723
|
| 243 |
+
2000 val perplexity 529.7121
|
| 244 |
+
2000 train 6.260824 (lr=2.2588e-03) (hash(x)=81308591)
|
| 245 |
+
2010 train 6.418819 (lr=2.2484e-03) (hash(x)=80205479)
|
| 246 |
+
2020 train 6.418782 (lr=2.2380e-03) (hash(x)=81008704)
|
| 247 |
+
2030 train 6.398265 (lr=2.2275e-03) (hash(x)=82914358)
|
| 248 |
+
2040 train 6.439533 (lr=2.2170e-03) (hash(x)=88064399)
|
| 249 |
+
2050 train 6.376195 (lr=2.2065e-03) (hash(x)=82889390)
|
| 250 |
+
2060 train 6.353879 (lr=2.1959e-03) (hash(x)=80913554)
|
| 251 |
+
2070 train 6.372523 (lr=2.1853e-03) (hash(x)=78121791)
|
| 252 |
+
2080 train 6.492655 (lr=2.1746e-03) (hash(x)=75356657)
|
| 253 |
+
2090 train 6.591593 (lr=2.1639e-03) (hash(x)=74048412)
|
| 254 |
+
2100 val loss 6.9126
|
| 255 |
+
2100 val perplexity 1004.8992
|
| 256 |
+
2100 train 6.747331 (lr=2.1532e-03) (hash(x)=68928225)
|
| 257 |
+
2110 train 6.814714 (lr=2.1424e-03) (hash(x)=93734745)
|
| 258 |
+
2120 train 6.702939 (lr=2.1316e-03) (hash(x)=72878164)
|
| 259 |
+
2130 train 6.678699 (lr=2.1208e-03) (hash(x)=71508226)
|
| 260 |
+
2140 train 7.058966 (lr=2.1099e-03) (hash(x)=84168671)
|
| 261 |
+
2150 train 6.827260 (lr=2.0990e-03) (hash(x)=74673639)
|
| 262 |
+
2160 train 6.843819 (lr=2.0881e-03) (hash(x)=75470031)
|
| 263 |
+
2170 train 6.816667 (lr=2.0771e-03) (hash(x)=74307890)
|
| 264 |
+
2180 train 6.746219 (lr=2.0661e-03) (hash(x)=77214245)
|
| 265 |
+
2190 train 6.676325 (lr=2.0550e-03) (hash(x)=88628359)
|
| 266 |
+
2200 val loss 6.9874
|
| 267 |
+
2200 val perplexity 1082.8878
|
| 268 |
+
2200 train 6.846850 (lr=2.0440e-03) (hash(x)=74779126)
|
| 269 |
+
2210 train 6.818467 (lr=2.0329e-03) (hash(x)=79817976)
|
| 270 |
+
2220 train 6.934974 (lr=2.0217e-03) (hash(x)=75258996)
|
| 271 |
+
2230 train 7.295659 (lr=2.0106e-03) (hash(x)=88164047)
|
| 272 |
+
2240 train 6.826830 (lr=1.9994e-03) (hash(x)=78580686)
|
| 273 |
+
2250 train 6.725634 (lr=1.9882e-03) (hash(x)=76510617)
|
| 274 |
+
2260 train 6.760932 (lr=1.9770e-03) (hash(x)=76043966)
|
| 275 |
+
2270 train 6.683221 (lr=1.9657e-03) (hash(x)=73062098)
|
| 276 |
+
2280 train 6.780467 (lr=1.9544e-03) (hash(x)=81885909)
|
| 277 |
+
2290 train 6.658337 (lr=1.9431e-03) (hash(x)=81481741)
|
| 278 |
+
2300 val loss 6.7018
|
| 279 |
+
2300 val perplexity 813.8480
|
| 280 |
+
2300 train 6.706089 (lr=1.9318e-03) (hash(x)=82104275)
|
| 281 |
+
2310 train 6.680802 (lr=1.9205e-03) (hash(x)=82137309)
|
| 282 |
+
2320 train 6.675594 (lr=1.9091e-03) (hash(x)=86741167)
|
| 283 |
+
2330 train 6.645253 (lr=1.8977e-03) (hash(x)=84796263)
|
| 284 |
+
2340 train 6.719346 (lr=1.8863e-03) (hash(x)=169966529)
|
| 285 |
+
2350 train 6.414638 (lr=1.8749e-03) (hash(x)=80006061)
|
| 286 |
+
2360 train 6.586424 (lr=1.8635e-03) (hash(x)=71090999)
|
| 287 |
+
2370 train 6.670510 (lr=1.8520e-03) (hash(x)=86641850)
|
| 288 |
+
2380 train 6.728713 (lr=1.8406e-03) (hash(x)=84573253)
|
| 289 |
+
2390 train 6.762067 (lr=1.8291e-03) (hash(x)=74483764)
|
| 290 |
+
2400 val loss 6.7124
|
| 291 |
+
2400 val perplexity 822.5165
|
| 292 |
+
2400 train 6.617848 (lr=1.8176e-03) (hash(x)=78327659)
|
| 293 |
+
2410 train 6.570099 (lr=1.8061e-03) (hash(x)=77273627)
|
| 294 |
+
2420 train 6.505845 (lr=1.7946e-03) (hash(x)=76938049)
|
| 295 |
+
2430 train 6.523451 (lr=1.7830e-03) (hash(x)=98449442)
|
| 296 |
+
2440 train 6.387842 (lr=1.7715e-03) (hash(x)=78157797)
|
| 297 |
+
2450 train 6.405251 (lr=1.7600e-03) (hash(x)=80637582)
|
| 298 |
+
2460 train 6.403747 (lr=1.7484e-03) (hash(x)=80023854)
|
| 299 |
+
2470 train 6.584394 (lr=1.7368e-03) (hash(x)=86678884)
|
| 300 |
+
2480 train 6.881507 (lr=1.7253e-03) (hash(x)=84086469)
|
| 301 |
+
2490 train 6.826900 (lr=1.7137e-03) (hash(x)=83131141)
|
| 302 |
+
2500 val loss 6.6460
|
| 303 |
+
2500 val perplexity 769.6839
|
| 304 |
+
2500 train 6.642836 (lr=1.7021e-03) (hash(x)=82583497)
|
| 305 |
+
2510 train 6.788582 (lr=1.6906e-03) (hash(x)=79164326)
|
| 306 |
+
2520 train 6.861624 (lr=1.6790e-03) (hash(x)=76456503)
|
| 307 |
+
2530 train 6.642696 (lr=1.6674e-03) (hash(x)=82001228)
|
| 308 |
+
2540 train 6.647473 (lr=1.6558e-03) (hash(x)=75615595)
|
| 309 |
+
2550 train 6.581892 (lr=1.6442e-03) (hash(x)=78397869)
|
| 310 |
+
2560 train 6.565693 (lr=1.6326e-03) (hash(x)=75547032)
|
| 311 |
+
2570 train 6.614096 (lr=1.6210e-03) (hash(x)=89201025)
|
| 312 |
+
2580 train 6.645628 (lr=1.6094e-03) (hash(x)=89856704)
|
| 313 |
+
2590 train 6.655800 (lr=1.5979e-03) (hash(x)=82175682)
|
| 314 |
+
2600 val loss 6.8106
|
| 315 |
+
2600 val perplexity 907.3824
|
| 316 |
+
2600 train 6.652885 (lr=1.5863e-03) (hash(x)=78312826)
|
| 317 |
+
2610 train 6.642991 (lr=1.5747e-03) (hash(x)=77066588)
|
| 318 |
+
2620 train 6.698375 (lr=1.5632e-03) (hash(x)=78666061)
|
| 319 |
+
2630 train 6.610326 (lr=1.5516e-03) (hash(x)=93762143)
|
| 320 |
+
2640 train 6.565598 (lr=1.5400e-03) (hash(x)=83191587)
|
| 321 |
+
2650 train 6.649939 (lr=1.5285e-03) (hash(x)=87169585)
|
| 322 |
+
2660 train 6.855712 (lr=1.5170e-03) (hash(x)=86426388)
|
| 323 |
+
2670 train 7.081698 (lr=1.5054e-03) (hash(x)=76692638)
|
| 324 |
+
2680 train 7.096290 (lr=1.4939e-03) (hash(x)=77446063)
|
| 325 |
+
2690 train 7.018864 (lr=1.4824e-03) (hash(x)=79809050)
|
| 326 |
+
2700 val loss 6.9486
|
| 327 |
+
2700 val perplexity 1041.6862
|
| 328 |
+
2700 train 6.886878 (lr=1.4709e-03) (hash(x)=83116823)
|
| 329 |
+
2710 train 6.885173 (lr=1.4594e-03) (hash(x)=75622148)
|
| 330 |
+
2720 train 6.790352 (lr=1.4480e-03) (hash(x)=80690512)
|
| 331 |
+
2730 train 6.853219 (lr=1.4365e-03) (hash(x)=78009984)
|
| 332 |
+
2740 train 6.908075 (lr=1.4251e-03) (hash(x)=74205488)
|
| 333 |
+
2750 train 6.855191 (lr=1.4137e-03) (hash(x)=91013332)
|
| 334 |
+
2760 train 6.769537 (lr=1.4023e-03) (hash(x)=76954961)
|
| 335 |
+
2770 train 6.915580 (lr=1.3909e-03) (hash(x)=79390317)
|
| 336 |
+
2780 train 6.805418 (lr=1.3795e-03) (hash(x)=70168783)
|
| 337 |
+
2790 train 6.816555 (lr=1.3682e-03) (hash(x)=76028417)
|
| 338 |
+
2800 val loss 6.9666
|
| 339 |
+
2800 val perplexity 1060.6105
|
| 340 |
+
2800 train 6.796491 (lr=1.3569e-03) (hash(x)=77656050)
|
| 341 |
+
2810 train 7.113094 (lr=1.3456e-03) (hash(x)=94248216)
|
| 342 |
+
2820 train 7.074017 (lr=1.3343e-03) (hash(x)=78305078)
|
| 343 |
+
2830 train 6.946254 (lr=1.3230e-03) (hash(x)=79948848)
|
| 344 |
+
2840 train 6.889913 (lr=1.3118e-03) (hash(x)=85341024)
|
| 345 |
+
2850 train 7.019835 (lr=1.3006e-03) (hash(x)=78735170)
|
| 346 |
+
2860 train 7.008215 (lr=1.2894e-03) (hash(x)=71616419)
|
| 347 |
+
2870 train 6.986315 (lr=1.2783e-03) (hash(x)=78656517)
|
| 348 |
+
2880 train 6.874069 (lr=1.2671e-03) (hash(x)=80073987)
|
| 349 |
+
2890 train 6.903335 (lr=1.2560e-03) (hash(x)=76894809)
|
| 350 |
+
2900 val loss 7.0127
|
| 351 |
+
2900 val perplexity 1110.6393
|
| 352 |
+
2900 train 6.981689 (lr=1.2450e-03) (hash(x)=80499838)
|
| 353 |
+
2910 train 6.875302 (lr=1.2339e-03) (hash(x)=72673354)
|
| 354 |
+
2920 train 6.885710 (lr=1.2229e-03) (hash(x)=84265768)
|
| 355 |
+
2930 train 7.137162 (lr=1.2119e-03) (hash(x)=79612060)
|
| 356 |
+
2940 train 6.993080 (lr=1.2010e-03) (hash(x)=74970087)
|
| 357 |
+
2950 train 7.047415 (lr=1.1901e-03) (hash(x)=84166818)
|
| 358 |
+
2960 train 6.940518 (lr=1.1792e-03) (hash(x)=89410221)
|
| 359 |
+
2970 train 6.896634 (lr=1.1684e-03) (hash(x)=75672566)
|
| 360 |
+
2980 train 6.753918 (lr=1.1576e-03) (hash(x)=81760314)
|
| 361 |
+
2990 train 6.846835 (lr=1.1468e-03) (hash(x)=80605200)
|
| 362 |
+
3000 val loss 6.9324
|
| 363 |
+
3000 val perplexity 1024.9341
|
| 364 |
+
3000 train 6.831442 (lr=1.1361e-03) (hash(x)=83804735)
|
| 365 |
+
3010 train 6.875754 (lr=1.1254e-03) (hash(x)=77015303)
|
| 366 |
+
3020 train 6.825679 (lr=1.1147e-03) (hash(x)=81464523)
|
| 367 |
+
3030 train 6.964807 (lr=1.1041e-03) (hash(x)=80168230)
|
| 368 |
+
3040 train 6.962859 (lr=1.0935e-03) (hash(x)=84817006)
|
| 369 |
+
3050 train 6.871988 (lr=1.0830e-03) (hash(x)=71601811)
|
| 370 |
+
3060 train 6.874843 (lr=1.0725e-03) (hash(x)=85499733)
|
| 371 |
+
3070 train 6.924728 (lr=1.0620e-03) (hash(x)=78670408)
|
| 372 |
+
3080 train 6.813707 (lr=1.0516e-03) (hash(x)=77120468)
|
| 373 |
+
3090 train 6.922147 (lr=1.0412e-03) (hash(x)=77927426)
|
| 374 |
+
3100 val loss 6.9421
|
| 375 |
+
3100 val perplexity 1034.9419
|
| 376 |
+
3100 train 6.830243 (lr=1.0309e-03) (hash(x)=83998606)
|
| 377 |
+
3110 train 6.848991 (lr=1.0207e-03) (hash(x)=77990218)
|
| 378 |
+
3120 train 6.830270 (lr=1.0104e-03) (hash(x)=81623970)
|
| 379 |
+
3130 train 6.753968 (lr=1.0002e-03) (hash(x)=75014781)
|
| 380 |
+
3140 train 6.792324 (lr=9.9011e-04) (hash(x)=72591250)
|
| 381 |
+
3150 train 6.854285 (lr=9.8002e-04) (hash(x)=81421847)
|
| 382 |
+
3160 train 6.878016 (lr=9.6999e-04) (hash(x)=75112631)
|
| 383 |
+
3170 train 7.018173 (lr=9.6000e-04) (hash(x)=87518033)
|
| 384 |
+
3180 train 7.093812 (lr=9.5007e-04) (hash(x)=78648348)
|
| 385 |
+
3190 train 6.966319 (lr=9.4019e-04) (hash(x)=78270029)
|
| 386 |
+
3200 val loss 7.0302
|
| 387 |
+
3200 val perplexity 1130.2474
|
| 388 |
+
3200 train 7.086451 (lr=9.3036e-04) (hash(x)=83380714)
|
| 389 |
+
3210 train 7.032772 (lr=9.2058e-04) (hash(x)=67804991)
|
| 390 |
+
3220 train 6.905227 (lr=9.1085e-04) (hash(x)=76968804)
|
| 391 |
+
3230 train 6.864734 (lr=9.0118e-04) (hash(x)=75798670)
|
| 392 |
+
3240 train 6.899312 (lr=8.9157e-04) (hash(x)=83192811)
|
| 393 |
+
3250 train 6.823362 (lr=8.8201e-04) (hash(x)=81659789)
|
| 394 |
+
3260 train 6.816227 (lr=8.7251e-04) (hash(x)=80622502)
|
| 395 |
+
3270 train 6.895934 (lr=8.6307e-04) (hash(x)=85436511)
|
| 396 |
+
3280 train 6.839864 (lr=8.5368e-04) (hash(x)=80589180)
|
| 397 |
+
3290 train 6.933308 (lr=8.4435e-04) (hash(x)=85747193)
|
| 398 |
+
3300 val loss 6.9504
|
| 399 |
+
3300 val perplexity 1043.5968
|
| 400 |
+
3300 train 6.895441 (lr=8.3508e-04) (hash(x)=74635692)
|
| 401 |
+
3310 train 6.971855 (lr=8.2588e-04) (hash(x)=82045455)
|
| 402 |
+
3320 train 6.804964 (lr=8.1673e-04) (hash(x)=73249173)
|
| 403 |
+
3330 train 6.845800 (lr=8.0764e-04) (hash(x)=71492338)
|
| 404 |
+
3340 train 6.940888 (lr=7.9862e-04) (hash(x)=76374071)
|
| 405 |
+
3350 train 6.893531 (lr=7.8966e-04) (hash(x)=80474064)
|
| 406 |
+
3360 train 6.865633 (lr=7.8076e-04) (hash(x)=84390892)
|
| 407 |
+
3370 train 6.943644 (lr=7.7192e-04) (hash(x)=83399949)
|
| 408 |
+
3380 train 6.941981 (lr=7.6315e-04) (hash(x)=77648059)
|
| 409 |
+
3390 train 7.001423 (lr=7.5445e-04) (hash(x)=80152701)
|
| 410 |
+
3400 val loss 6.8738
|
| 411 |
+
3400 val perplexity 966.6215
|
| 412 |
+
3400 train 6.942432 (lr=7.4581e-04) (hash(x)=81952545)
|
| 413 |
+
3410 train 6.910998 (lr=7.3724e-04) (hash(x)=80908993)
|
| 414 |
+
3420 train 6.839645 (lr=7.2874e-04) (hash(x)=79610037)
|
| 415 |
+
3430 train 6.861483 (lr=7.2030e-04) (hash(x)=87624382)
|
| 416 |
+
3440 train 6.800820 (lr=7.1193e-04) (hash(x)=82336381)
|
| 417 |
+
3450 train 6.820130 (lr=7.0363e-04) (hash(x)=87687835)
|
| 418 |
+
3460 train 6.915255 (lr=6.9541e-04) (hash(x)=76376135)
|
| 419 |
+
3470 train 6.844012 (lr=6.8725e-04) (hash(x)=79331391)
|
| 420 |
+
3480 train 6.887217 (lr=6.7916e-04) (hash(x)=107002681)
|
| 421 |
+
3490 train 6.788852 (lr=6.7114e-04) (hash(x)=77465514)
|
| 422 |
+
3500 val loss 6.9214
|
| 423 |
+
3500 val perplexity 1013.7369
|
| 424 |
+
3500 train 6.862830 (lr=6.6320e-04) (hash(x)=88237229)
|
| 425 |
+
3510 train 6.979153 (lr=6.5533e-04) (hash(x)=81011739)
|
| 426 |
+
3520 train 6.639634 (lr=6.4753e-04) (hash(x)=64643427)
|
| 427 |
+
3530 train 6.911961 (lr=6.3981e-04) (hash(x)=78029539)
|
| 428 |
+
3540 train 7.047397 (lr=6.3216e-04) (hash(x)=83188968)
|
| 429 |
+
3550 train 6.870165 (lr=6.2458e-04) (hash(x)=86104185)
|
| 430 |
+
3560 train 6.950630 (lr=6.1708e-04) (hash(x)=81993629)
|
| 431 |
+
3570 train 6.851932 (lr=6.0966e-04) (hash(x)=72719368)
|
| 432 |
+
3580 train 6.872307 (lr=6.0231e-04) (hash(x)=90091487)
|
| 433 |
+
3590 train 6.808402 (lr=5.9504e-04) (hash(x)=77393152)
|
| 434 |
+
3600 val loss 6.8832
|
| 435 |
+
3600 val perplexity 975.7537
|
| 436 |
+
3600 train 6.843101 (lr=5.8785e-04) (hash(x)=73103504)
|
| 437 |
+
3610 train 6.895177 (lr=5.8074e-04) (hash(x)=84429400)
|
| 438 |
+
3620 train 6.829403 (lr=5.7370e-04) (hash(x)=77021795)
|
| 439 |
+
3630 train 6.882319 (lr=5.6675e-04) (hash(x)=92830605)
|
| 440 |
+
3640 train 6.856334 (lr=5.5987e-04) (hash(x)=78313175)
|
| 441 |
+
3650 train 6.826724 (lr=5.5308e-04) (hash(x)=85395549)
|
| 442 |
+
3660 train 6.630293 (lr=5.4636e-04) (hash(x)=78114459)
|
| 443 |
+
3670 train 6.964176 (lr=5.3973e-04) (hash(x)=74968316)
|
| 444 |
+
3680 train 6.855954 (lr=5.3318e-04) (hash(x)=73358737)
|
| 445 |
+
3690 train 7.165024 (lr=5.2671e-04) (hash(x)=76399442)
|
| 446 |
+
3700 val loss 7.0451
|
| 447 |
+
3700 val perplexity 1147.2008
|
| 448 |
+
3700 train 6.889978 (lr=5.2033e-04) (hash(x)=74039273)
|
| 449 |
+
3710 train 7.001658 (lr=5.1402e-04) (hash(x)=79299680)
|
| 450 |
+
3720 train 6.987977 (lr=5.0780e-04) (hash(x)=94701498)
|
| 451 |
+
3730 train 6.949964 (lr=5.0167e-04) (hash(x)=75352071)
|
| 452 |
+
3740 train 6.916829 (lr=4.9562e-04) (hash(x)=77909487)
|
| 453 |
+
3750 train 6.885458 (lr=4.8965e-04) (hash(x)=73986730)
|
| 454 |
+
3760 train 6.977623 (lr=4.8377e-04) (hash(x)=79325763)
|
| 455 |
+
3770 train 6.835709 (lr=4.7798e-04) (hash(x)=72457818)
|
| 456 |
+
3780 train 7.000257 (lr=4.7227e-04) (hash(x)=71775590)
|
| 457 |
+
3790 train 6.972394 (lr=4.6665e-04) (hash(x)=82638943)
|
| 458 |
+
3800 val loss 7.0437
|
| 459 |
+
3800 val perplexity 1145.6364
|
| 460 |
+
3800 train 6.952043 (lr=4.6112e-04) (hash(x)=79965893)
|
| 461 |
+
3810 train 6.859846 (lr=4.5567e-04) (hash(x)=72598235)
|
| 462 |
+
3820 train 6.960604 (lr=4.5031e-04) (hash(x)=83113889)
|
| 463 |
+
3830 train 6.975122 (lr=4.4504e-04) (hash(x)=74434590)
|
| 464 |
+
3840 train 6.941326 (lr=4.3986e-04) (hash(x)=82860348)
|
| 465 |
+
3850 train 6.834870 (lr=4.3477e-04) (hash(x)=78067565)
|
| 466 |
+
3860 train 6.837601 (lr=4.2977e-04) (hash(x)=82592498)
|
| 467 |
+
3870 train 6.879895 (lr=4.2486e-04) (hash(x)=81820733)
|
| 468 |
+
3880 train 6.876242 (lr=4.2004e-04) (hash(x)=87709040)
|
| 469 |
+
3890 train 7.083454 (lr=4.1530e-04) (hash(x)=70379093)
|
| 470 |
+
3900 val loss 6.9546
|
| 471 |
+
3900 val perplexity 1047.9980
|
| 472 |
+
3900 train 6.950953 (lr=4.1066e-04) (hash(x)=76597431)
|
| 473 |
+
3910 train 6.983864 (lr=4.0611e-04) (hash(x)=90490716)
|
| 474 |
+
3920 train 6.862394 (lr=4.0166e-04) (hash(x)=81970659)
|
| 475 |
+
3930 train 6.884658 (lr=3.9729e-04) (hash(x)=81496334)
|
| 476 |
+
3940 train 6.778016 (lr=3.9302e-04) (hash(x)=75717605)
|
| 477 |
+
3950 train 6.833332 (lr=3.8884e-04) (hash(x)=82749357)
|
| 478 |
+
3960 train 6.848589 (lr=3.8475e-04) (hash(x)=89355157)
|
| 479 |
+
3970 train 6.754154 (lr=3.8076e-04) (hash(x)=78980403)
|
| 480 |
+
3980 train 6.756855 (lr=3.7685e-04) (hash(x)=76627217)
|
| 481 |
+
3990 train 6.766465 (lr=3.7305e-04) (hash(x)=72412879)
|
| 482 |
+
4000 val loss 6.8247
|
| 483 |
+
4000 val perplexity 920.3120
|
| 484 |
+
4000 train 6.822633 (lr=3.6933e-04) (hash(x)=83018142)
|
| 485 |
+
4010 train 6.704081 (lr=3.6572e-04) (hash(x)=81272436)
|
| 486 |
+
4020 train 6.757352 (lr=3.6219e-04) (hash(x)=85497482)
|
| 487 |
+
4030 train 6.913620 (lr=3.5876e-04) (hash(x)=85241734)
|
| 488 |
+
4040 train 6.841537 (lr=3.5543e-04) (hash(x)=77925307)
|
| 489 |
+
4050 train 6.838606 (lr=3.5219e-04) (hash(x)=84826179)
|
| 490 |
+
4060 train 6.888925 (lr=3.4905e-04) (hash(x)=83606764)
|
| 491 |
+
4070 train 6.861482 (lr=3.4600e-04) (hash(x)=80567590)
|
| 492 |
+
4080 train 6.779081 (lr=3.4305e-04) (hash(x)=76860998)
|
| 493 |
+
4090 train 6.718825 (lr=3.4019e-04) (hash(x)=74902328)
|
| 494 |
+
4100 val loss 6.8252
|
| 495 |
+
4100 val perplexity 920.7927
|
| 496 |
+
4100 train 6.812160 (lr=3.3744e-04) (hash(x)=82832041)
|
| 497 |
+
4110 train 6.796019 (lr=3.3477e-04) (hash(x)=79143262)
|
| 498 |
+
4120 train 6.780811 (lr=3.3221e-04) (hash(x)=77038149)
|
| 499 |
+
4130 train 6.862268 (lr=3.2974e-04) (hash(x)=86339074)
|
| 500 |
+
4140 train 6.750194 (lr=3.2737e-04) (hash(x)=76686216)
|
| 501 |
+
4150 train 6.800090 (lr=3.2510e-04) (hash(x)=70522682)
|
| 502 |
+
4160 train 6.824504 (lr=3.2292e-04) (hash(x)=90958555)
|
| 503 |
+
4170 train 6.836328 (lr=3.2085e-04) (hash(x)=91463532)
|
| 504 |
+
4180 train 6.768838 (lr=3.1887e-04) (hash(x)=81959329)
|
| 505 |
+
4190 train 6.774848 (lr=3.1699e-04) (hash(x)=83146752)
|
| 506 |
+
4200 val loss 6.7896
|
| 507 |
+
4200 val perplexity 888.5279
|
| 508 |
+
4200 train 6.723736 (lr=3.1520e-04) (hash(x)=78361715)
|
| 509 |
+
4210 train 6.793519 (lr=3.1352e-04) (hash(x)=87364889)
|
| 510 |
+
4220 train 6.670817 (lr=3.1193e-04) (hash(x)=70465156)
|
| 511 |
+
4230 train 6.803814 (lr=3.1044e-04) (hash(x)=84524081)
|
| 512 |
+
4240 train 6.801944 (lr=3.0905e-04) (hash(x)=77824868)
|
| 513 |
+
4250 train 6.822649 (lr=3.0776e-04) (hash(x)=81710711)
|
| 514 |
+
4260 train 6.755538 (lr=3.0657e-04) (hash(x)=76362728)
|
| 515 |
+
4270 train 6.731719 (lr=3.0548e-04) (hash(x)=83115208)
|
| 516 |
+
4280 train 6.759505 (lr=3.0449e-04) (hash(x)=87218314)
|
| 517 |
+
4290 train 6.824706 (lr=3.0359e-04) (hash(x)=74582673)
|
| 518 |
+
4300 val loss 6.8751
|
| 519 |
+
4300 val perplexity 967.8686
|
| 520 |
+
4300 train 6.848835 (lr=3.0280e-04) (hash(x)=77379615)
|
| 521 |
+
4310 train 6.750901 (lr=3.0210e-04) (hash(x)=78669579)
|
| 522 |
+
4320 train 6.815701 (lr=3.0150e-04) (hash(x)=83066608)
|
| 523 |
+
4330 train 6.841862 (lr=3.0101e-04) (hash(x)=83037340)
|
| 524 |
+
4340 train 6.940447 (lr=3.0061e-04) (hash(x)=82849771)
|
| 525 |
+
4350 train 6.866872 (lr=3.0031e-04) (hash(x)=76693985)
|
| 526 |
+
4360 train 6.843583 (lr=3.0011e-04) (hash(x)=77745394)
|
| 527 |
+
4370 train 6.785136 (lr=3.0001e-04) (hash(x)=79954388)
|
| 528 |
+
4374 val loss 6.8611
|
| 529 |
+
4374 val perplexity 954.4183
|
12_mini_head_one_mask_per_head_1_latent_mask/model_04374.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 97706354
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e4b9f4b08ae0c1aa234c7d457e0b6242ffd74d7fd0b4fc2d053dde9c753d545
|
| 3 |
size 97706354
|
12_mini_head_one_mask_per_head_1_latent_mask/optimizer_04374.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 189135414
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36d10aa8b7a52812cfec762b9b467aff8091c795a1eaddfb47d7df2e2780f77b
|
| 3 |
size 189135414
|