Update tokenizer_config.json
Browse files- tokenizer_config.json +342 -15
tokenizer_config.json
CHANGED
@@ -34,7 +34,7 @@
|
|
34 |
"special": false
|
35 |
},
|
36 |
"50256": {
|
37 |
-
"content": "
|
38 |
"lstrip": false,
|
39 |
"normalized": false,
|
40 |
"rstrip": false,
|
@@ -42,7 +42,7 @@
|
|
42 |
"special": true
|
43 |
},
|
44 |
"50257": {
|
45 |
-
"content": "
|
46 |
"lstrip": false,
|
47 |
"normalized": false,
|
48 |
"rstrip": false,
|
@@ -266,36 +266,340 @@
|
|
266 |
"special": false
|
267 |
},
|
268 |
"50285": {
|
269 |
-
"content": "
|
270 |
"lstrip": false,
|
271 |
-
"normalized":
|
272 |
"rstrip": false,
|
273 |
"single_word": false,
|
274 |
-
"special":
|
275 |
},
|
276 |
"50286": {
|
277 |
-
"content": "
|
278 |
"lstrip": false,
|
279 |
-
"normalized":
|
280 |
"rstrip": false,
|
281 |
"single_word": false,
|
282 |
-
"special":
|
283 |
},
|
284 |
"50287": {
|
285 |
-
"content": "
|
286 |
"lstrip": false,
|
287 |
-
"normalized":
|
288 |
"rstrip": false,
|
289 |
"single_word": false,
|
290 |
-
"special":
|
291 |
},
|
292 |
"50288": {
|
293 |
-
"content": "
|
294 |
"lstrip": false,
|
295 |
-
"normalized":
|
296 |
"rstrip": false,
|
297 |
"single_word": false,
|
298 |
-
"special":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
299 |
}
|
300 |
},
|
301 |
"additional_special_tokens": [
|
@@ -325,5 +629,28 @@
|
|
325 |
"unk_token": "<unk>",
|
326 |
"use_auth_token": true,
|
327 |
"use_fast_tokenizer": true,
|
328 |
-
"wordpieces_prefix": "##"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
329 |
}
|
|
|
34 |
"special": false
|
35 |
},
|
36 |
"50256": {
|
37 |
+
"content": "",
|
38 |
"lstrip": false,
|
39 |
"normalized": false,
|
40 |
"rstrip": false,
|
|
|
42 |
"special": true
|
43 |
},
|
44 |
"50257": {
|
45 |
+
"content": "",
|
46 |
"lstrip": false,
|
47 |
"normalized": false,
|
48 |
"rstrip": false,
|
|
|
266 |
"special": false
|
267 |
},
|
268 |
"50285": {
|
269 |
+
"content": "active_listener",
|
270 |
"lstrip": false,
|
271 |
+
"normalized": true,
|
272 |
"rstrip": false,
|
273 |
"single_word": false,
|
274 |
+
"special": false
|
275 |
},
|
276 |
"50286": {
|
277 |
+
"content": "emotional_intelligence",
|
278 |
"lstrip": false,
|
279 |
+
"normalized": true,
|
280 |
"rstrip": false,
|
281 |
"single_word": false,
|
282 |
+
"special": false
|
283 |
},
|
284 |
"50287": {
|
285 |
+
"content": "interpersonal_skill",
|
286 |
"lstrip": false,
|
287 |
+
"normalized": true,
|
288 |
"rstrip": false,
|
289 |
"single_word": false,
|
290 |
+
"special": false
|
291 |
},
|
292 |
"50288": {
|
293 |
+
"content": "non_verbal_cue",
|
294 |
"lstrip": false,
|
295 |
+
"normalized": true,
|
296 |
"rstrip": false,
|
297 |
"single_word": false,
|
298 |
+
"special": false
|
299 |
+
},
|
300 |
+
"50289": {
|
301 |
+
"content": "personal_boundaries",
|
302 |
+
"lstrip": false,
|
303 |
+
"normalized": true,
|
304 |
+
"rstrip": false,
|
305 |
+
"single_word": false,
|
306 |
+
"special": false
|
307 |
+
},
|
308 |
+
"50290": {
|
309 |
+
"content": "active_listener",
|
310 |
+
"lstrip": false,
|
311 |
+
"normalized": true,
|
312 |
+
"rstrip": false,
|
313 |
+
"single_word": false,
|
314 |
+
"special": false
|
315 |
+
},
|
316 |
+
"50291": {
|
317 |
+
"content": "non_judgmental",
|
318 |
+
"lstrip": false,
|
319 |
+
"normalized": true,
|
320 |
+
"rstrip": false,
|
321 |
+
"single_word": false,
|
322 |
+
"special": false
|
323 |
+
},
|
324 |
+
"50292": {
|
325 |
+
"content": "cultural_sensitivity",
|
326 |
+
"lstrip": false,
|
327 |
+
"normalized": true,
|
328 |
+
"rstrip": false,
|
329 |
+
"single_word": false,
|
330 |
+
"special": false
|
331 |
+
},
|
332 |
+
"50293": {
|
333 |
+
"content": "emotional_regulation",
|
334 |
+
"lstrip": false,
|
335 |
+
"normalized": true,
|
336 |
+
"rstrip": false,
|
337 |
+
"single_word": false,
|
338 |
+
"special": false
|
339 |
+
},
|
340 |
+
"50294": {
|
341 |
+
"content": "self_awareness",
|
342 |
+
"lstrip": false,
|
343 |
+
"normalized": true,
|
344 |
+
"rstrip": false,
|
345 |
+
"single_word": false,
|
346 |
+
"special": false
|
347 |
+
},
|
348 |
+
"50295": {
|
349 |
+
"content": "social_skill",
|
350 |
+
"lstrip": false,
|
351 |
+
"normalized": true,
|
352 |
+
"rstrip": false,
|
353 |
+
"single_word": false,
|
354 |
+
"special": false
|
355 |
+
},
|
356 |
+
"50296": {
|
357 |
+
"content": "communication_style",
|
358 |
+
"lstrip": false,
|
359 |
+
"normalized": true,
|
360 |
+
"rstrip": false,
|
361 |
+
"single_word": false,
|
362 |
+
"special": false
|
363 |
+
},
|
364 |
+
"50297": {
|
365 |
+
"content": "conversational_skill",
|
366 |
+
"lstrip": false,
|
367 |
+
"normalized": true,
|
368 |
+
"rstrip": false,
|
369 |
+
"single_word": false,
|
370 |
+
"special": false
|
371 |
+
},
|
372 |
+
"50298": {
|
373 |
+
"content": "emotional_literacy",
|
374 |
+
"lstrip": false,
|
375 |
+
"normalized": true,
|
376 |
+
"rstrip": false,
|
377 |
+
"single_word": false,
|
378 |
+
"special": false
|
379 |
+
},
|
380 |
+
"50299": {
|
381 |
+
"content": "interpersonal_communication",
|
382 |
+
"lstrip": false,
|
383 |
+
"normalized": true,
|
384 |
+
"rstrip": false,
|
385 |
+
"single_word": false,
|
386 |
+
"special": false
|
387 |
+
},
|
388 |
+
"50300": {
|
389 |
+
"content": "resilience",
|
390 |
+
"lstrip": false,
|
391 |
+
"normalized": true,
|
392 |
+
"rstrip": false,
|
393 |
+
"single_word": false,
|
394 |
+
"special": false
|
395 |
+
},
|
396 |
+
"50301": {
|
397 |
+
"content": "stress_management",
|
398 |
+
"lstrip": false,
|
399 |
+
"normalized": true,
|
400 |
+
"rstrip": false,
|
401 |
+
"single_word": false,
|
402 |
+
"special": false
|
403 |
+
},
|
404 |
+
"50302": {
|
405 |
+
"content": "self_management",
|
406 |
+
"lstrip": false,
|
407 |
+
"normalized": true,
|
408 |
+
"rstrip": false,
|
409 |
+
"single_word": false,
|
410 |
+
"special": false
|
411 |
+
},
|
412 |
+
"50303": {
|
413 |
+
"content": "self_efficacy",
|
414 |
+
"lstrip": false,
|
415 |
+
"normalized": true,
|
416 |
+
"rstrip": false,
|
417 |
+
"single_word": false,
|
418 |
+
"special": false
|
419 |
+
},
|
420 |
+
"50304": {
|
421 |
+
"content": "self_motivation",
|
422 |
+
"lstrip": false,
|
423 |
+
"normalized": true,
|
424 |
+
"rstrip": false,
|
425 |
+
"single_word": false,
|
426 |
+
"special": false
|
427 |
+
},
|
428 |
+
"50305": {
|
429 |
+
"content": "self_control",
|
430 |
+
"lstrip": false,
|
431 |
+
"normalized": true,
|
432 |
+
"rstrip": false,
|
433 |
+
"single_word": false,
|
434 |
+
"special": false
|
435 |
+
},
|
436 |
+
"50306": {
|
437 |
+
"content": "self_esteem",
|
438 |
+
"lstrip": false,
|
439 |
+
"normalized": true,
|
440 |
+
"rstrip": false,
|
441 |
+
"single_word": false,
|
442 |
+
"special": false
|
443 |
+
},
|
444 |
+
"50307": {
|
445 |
+
"content": "emotional_balance",
|
446 |
+
"lstrip": false,
|
447 |
+
"normalized": true,
|
448 |
+
"rstrip": false,
|
449 |
+
"single_word": false,
|
450 |
+
"special": false
|
451 |
+
},
|
452 |
+
"50308": {
|
453 |
+
"content": "stress_reduction",
|
454 |
+
"lstrip": false,
|
455 |
+
"normalized": true,
|
456 |
+
"rstrip": false,
|
457 |
+
"single_word": false,
|
458 |
+
"special": false
|
459 |
+
},
|
460 |
+
"50309": {
|
461 |
+
"content": "goal_setting",
|
462 |
+
"lstrip": false,
|
463 |
+
"normalized": true,
|
464 |
+
"rstrip": false,
|
465 |
+
"single_word": false,
|
466 |
+
"special": false
|
467 |
+
},
|
468 |
+
"50310": {
|
469 |
+
"content": "time_management",
|
470 |
+
"lstrip": false,
|
471 |
+
"normalized": true,
|
472 |
+
"rstrip": false,
|
473 |
+
"single_word": false,
|
474 |
+
"special": false
|
475 |
+
},
|
476 |
+
"50311": {
|
477 |
+
"content": "organization_skill",
|
478 |
+
"lstrip": false,
|
479 |
+
"normalized": true,
|
480 |
+
"rstrip": false,
|
481 |
+
"single_word": false,
|
482 |
+
"special": false
|
483 |
+
},
|
484 |
+
"50312": {
|
485 |
+
"content": "self_directed_learning",
|
486 |
+
"lstrip": false,
|
487 |
+
"normalized": true,
|
488 |
+
"rstrip": false,
|
489 |
+
"single_word": false,
|
490 |
+
"special": false
|
491 |
+
},
|
492 |
+
"50313": {
|
493 |
+
"content": "self_reflection",
|
494 |
+
"lstrip": false,
|
495 |
+
"normalized": true,
|
496 |
+
"rstrip": false,
|
497 |
+
"single_word": false,
|
498 |
+
"special": false
|
499 |
+
},
|
500 |
+
"50314": {
|
501 |
+
"content": "empathy_map",
|
502 |
+
"lstrip": false,
|
503 |
+
"normalized": true,
|
504 |
+
"rstrip": false,
|
505 |
+
"single_word": false,
|
506 |
+
"special": false
|
507 |
+
},
|
508 |
+
"50315": {
|
509 |
+
"content": "interpersonal_relationship",
|
510 |
+
"lstrip": false,
|
511 |
+
"normalized": true,
|
512 |
+
"rstrip": false,
|
513 |
+
"single_word": false,
|
514 |
+
"special": false
|
515 |
+
},
|
516 |
+
"50316": {
|
517 |
+
"content": "personal_growth",
|
518 |
+
"lstrip": false,
|
519 |
+
"normalized": true,
|
520 |
+
"rstrip": false,
|
521 |
+
"single_word": false,
|
522 |
+
"special": false
|
523 |
+
},
|
524 |
+
"50317": {
|
525 |
+
"content": "personal_development",
|
526 |
+
"lstrip": false,
|
527 |
+
"normalized": true,
|
528 |
+
"rstrip": false,
|
529 |
+
"single_word": false,
|
530 |
+
"special": false
|
531 |
+
},
|
532 |
+
"50318": {
|
533 |
+
"content": "mindfulness",
|
534 |
+
"lstrip": false,
|
535 |
+
"normalized": true,
|
536 |
+
"rstrip": false,
|
537 |
+
"single_word": false,
|
538 |
+
"special": false
|
539 |
+
},
|
540 |
+
"50319": {
|
541 |
+
"content": "personal_well_being",
|
542 |
+
"lstrip": false,
|
543 |
+
"normalized": true,
|
544 |
+
"rstrip": false,
|
545 |
+
"single_word": false,
|
546 |
+
"special": false
|
547 |
+
},
|
548 |
+
"50320": {
|
549 |
+
"content": "personal_resilience",
|
550 |
+
"lstrip": false,
|
551 |
+
"normalized": true,
|
552 |
+
"rstrip": false,
|
553 |
+
"single_word": false,
|
554 |
+
"special": false
|
555 |
+
},
|
556 |
+
"50321": {
|
557 |
+
"content": "self_care",
|
558 |
+
"lstrip": false,
|
559 |
+
"normalized": true,
|
560 |
+
"rstrip": false,
|
561 |
+
"single_word": false,
|
562 |
+
"special": false
|
563 |
+
},
|
564 |
+
"50322": {
|
565 |
+
"content": "self_improvement",
|
566 |
+
"lstrip": false,
|
567 |
+
"normalized": true,
|
568 |
+
"rstrip": false,
|
569 |
+
"single_word": false,
|
570 |
+
"special": false
|
571 |
+
},
|
572 |
+
"50323": {
|
573 |
+
"content": "personal_growth_plan",
|
574 |
+
"lstrip": false,
|
575 |
+
"normalized": true,
|
576 |
+
"rstrip": false,
|
577 |
+
"single_word": false,
|
578 |
+
"special": false
|
579 |
+
},
|
580 |
+
"50324": {
|
581 |
+
"content": "life_skill",
|
582 |
+
"lstrip": false,
|
583 |
+
"normalized": true,
|
584 |
+
"rstrip": false,
|
585 |
+
"single_word": false,
|
586 |
+
"special": false
|
587 |
+
},
|
588 |
+
"50325": {
|
589 |
+
"content": "self_renewal",
|
590 |
+
"lstrip": false,
|
591 |
+
"normalized": true,
|
592 |
+
"rstrip": false,
|
593 |
+
"single_word": false,
|
594 |
+
"special": false
|
595 |
+
},
|
596 |
+
"50326": {
|
597 |
+
"content": "self_growth",
|
598 |
+
"lstrip": false,
|
599 |
+
"normalized": true,
|
600 |
+
"rstrip": false,
|
601 |
+
"single_word": false,
|
602 |
+
"special": false
|
603 |
}
|
604 |
},
|
605 |
"additional_special_tokens": [
|
|
|
629 |
"unk_token": "<unk>",
|
630 |
"use_auth_token": true,
|
631 |
"use_fast_tokenizer": true,
|
632 |
+
"wordpieces_prefix": "##",
|
633 |
+
"enable_sentencepiece": true,
|
634 |
+
"bpe_dropout": 0.1,
|
635 |
+
"interleaved_tokenization": true,
|
636 |
+
"sentence_piece_model_path": "path/to/sentencepiece.model",
|
637 |
+
"additional_vocab_files": {
|
638 |
+
"bpe_vocab_file": "path/to/bpe_vocab.json",
|
639 |
+
"bpe_merges_file": "path/to/bpe_merges.txt"
|
640 |
+
},
|
641 |
+
"advanced_options": {
|
642 |
+
"dynamic_padding": true,
|
643 |
+
"lexical_analysis": {
|
644 |
+
"enable": true,
|
645 |
+
"methods": ["morphological", "syntactic", "semantic"]
|
646 |
+
},
|
647 |
+
"subword_regularization": {
|
648 |
+
"enable": true,
|
649 |
+
"techniques": ["sampling", "dropping"]
|
650 |
+
},
|
651 |
+
"language_model_integration": {
|
652 |
+
"enable": true,
|
653 |
+
"models": ["bert-base-uncased", "roberta-base"]
|
654 |
+
}
|
655 |
+
}
|
656 |
}
|