End of training
Browse files- README.md +4 -4
- config.json +1 -1
- generation_config.json +1 -1
- pytorch_model.bin +2 -2
- tokenizer.json +204 -204
- tokenizer_config.json +204 -204
- training_args.bin +2 -2
README.md
CHANGED
@@ -39,8 +39,8 @@ The following hyperparameters were used during training:
|
|
39 |
- total_train_batch_size: 20
|
40 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
41 |
- lr_scheduler_type: linear
|
42 |
-
- lr_scheduler_warmup_steps:
|
43 |
-
- num_epochs:
|
44 |
|
45 |
### Training results
|
46 |
|
@@ -48,7 +48,7 @@ The following hyperparameters were used during training:
|
|
48 |
|
49 |
### Framework versions
|
50 |
|
51 |
-
- Transformers 4.34.
|
52 |
-
- Pytorch 2.0
|
53 |
- Datasets 2.14.5
|
54 |
- Tokenizers 0.14.1
|
|
|
39 |
- total_train_batch_size: 20
|
40 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
41 |
- lr_scheduler_type: linear
|
42 |
+
- lr_scheduler_warmup_steps: 550
|
43 |
+
- num_epochs: 4
|
44 |
|
45 |
### Training results
|
46 |
|
|
|
48 |
|
49 |
### Framework versions
|
50 |
|
51 |
+
- Transformers 4.34.1
|
52 |
+
- Pytorch 2.1.0+cu118
|
53 |
- Datasets 2.14.5
|
54 |
- Tokenizers 0.14.1
|
config.json
CHANGED
@@ -49,7 +49,7 @@
|
|
49 |
"scale_embedding": true,
|
50 |
"static_position_embeddings": true,
|
51 |
"torch_dtype": "float32",
|
52 |
-
"transformers_version": "4.34.
|
53 |
"use_cache": true,
|
54 |
"vocab_size": 96103
|
55 |
}
|
|
|
49 |
"scale_embedding": true,
|
50 |
"static_position_embeddings": true,
|
51 |
"torch_dtype": "float32",
|
52 |
+
"transformers_version": "4.34.1",
|
53 |
"use_cache": true,
|
54 |
"vocab_size": 96103
|
55 |
}
|
generation_config.json
CHANGED
@@ -9,5 +9,5 @@
|
|
9 |
"min_length": 32,
|
10 |
"num_beams": 8,
|
11 |
"pad_token_id": 0,
|
12 |
-
"transformers_version": "4.34.
|
13 |
}
|
|
|
9 |
"min_length": 32,
|
10 |
"num_beams": 8,
|
11 |
"pad_token_id": 0,
|
12 |
+
"transformers_version": "4.34.1"
|
13 |
}
|
pytorch_model.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 sha256:3ead3e0d65f1362d18e5a6575f51c3ac1c88679ec136a4f998c0dbd67d5fbb32
|
3 |
+
size 2283805098
|
tokenizer.json
CHANGED
@@ -39,8 +39,8 @@
|
|
39 |
"id": 2,
|
40 |
"content": "<mask_1>",
|
41 |
"single_word": false,
|
42 |
-
"lstrip":
|
43 |
-
"rstrip":
|
44 |
"normalized": false,
|
45 |
"special": true
|
46 |
},
|
@@ -57,8 +57,8 @@
|
|
57 |
"id": 4,
|
58 |
"content": "<unk_2>",
|
59 |
"single_word": false,
|
60 |
-
"lstrip":
|
61 |
-
"rstrip":
|
62 |
"normalized": false,
|
63 |
"special": true
|
64 |
},
|
@@ -66,8 +66,8 @@
|
|
66 |
"id": 5,
|
67 |
"content": "<unk_3>",
|
68 |
"single_word": false,
|
69 |
-
"lstrip":
|
70 |
-
"rstrip":
|
71 |
"normalized": false,
|
72 |
"special": true
|
73 |
},
|
@@ -75,8 +75,8 @@
|
|
75 |
"id": 6,
|
76 |
"content": "<unk_4>",
|
77 |
"single_word": false,
|
78 |
-
"lstrip":
|
79 |
-
"rstrip":
|
80 |
"normalized": false,
|
81 |
"special": true
|
82 |
},
|
@@ -84,8 +84,8 @@
|
|
84 |
"id": 7,
|
85 |
"content": "<unk_5>",
|
86 |
"single_word": false,
|
87 |
-
"lstrip":
|
88 |
-
"rstrip":
|
89 |
"normalized": false,
|
90 |
"special": true
|
91 |
},
|
@@ -93,8 +93,8 @@
|
|
93 |
"id": 8,
|
94 |
"content": "<unk_6>",
|
95 |
"single_word": false,
|
96 |
-
"lstrip":
|
97 |
-
"rstrip":
|
98 |
"normalized": false,
|
99 |
"special": true
|
100 |
},
|
@@ -102,8 +102,8 @@
|
|
102 |
"id": 9,
|
103 |
"content": "<unk_7>",
|
104 |
"single_word": false,
|
105 |
-
"lstrip":
|
106 |
-
"rstrip":
|
107 |
"normalized": false,
|
108 |
"special": true
|
109 |
},
|
@@ -111,8 +111,8 @@
|
|
111 |
"id": 10,
|
112 |
"content": "<unk_8>",
|
113 |
"single_word": false,
|
114 |
-
"lstrip":
|
115 |
-
"rstrip":
|
116 |
"normalized": false,
|
117 |
"special": true
|
118 |
},
|
@@ -120,8 +120,8 @@
|
|
120 |
"id": 11,
|
121 |
"content": "<unk_9>",
|
122 |
"single_word": false,
|
123 |
-
"lstrip":
|
124 |
-
"rstrip":
|
125 |
"normalized": false,
|
126 |
"special": true
|
127 |
},
|
@@ -129,8 +129,8 @@
|
|
129 |
"id": 12,
|
130 |
"content": "<unk_10>",
|
131 |
"single_word": false,
|
132 |
-
"lstrip":
|
133 |
-
"rstrip":
|
134 |
"normalized": false,
|
135 |
"special": true
|
136 |
},
|
@@ -138,8 +138,8 @@
|
|
138 |
"id": 13,
|
139 |
"content": "<unk_11>",
|
140 |
"single_word": false,
|
141 |
-
"lstrip":
|
142 |
-
"rstrip":
|
143 |
"normalized": false,
|
144 |
"special": true
|
145 |
},
|
@@ -147,8 +147,8 @@
|
|
147 |
"id": 14,
|
148 |
"content": "<unk_12>",
|
149 |
"single_word": false,
|
150 |
-
"lstrip":
|
151 |
-
"rstrip":
|
152 |
"normalized": false,
|
153 |
"special": true
|
154 |
},
|
@@ -156,8 +156,8 @@
|
|
156 |
"id": 15,
|
157 |
"content": "<unk_13>",
|
158 |
"single_word": false,
|
159 |
-
"lstrip":
|
160 |
-
"rstrip":
|
161 |
"normalized": false,
|
162 |
"special": true
|
163 |
},
|
@@ -165,8 +165,8 @@
|
|
165 |
"id": 16,
|
166 |
"content": "<unk_14>",
|
167 |
"single_word": false,
|
168 |
-
"lstrip":
|
169 |
-
"rstrip":
|
170 |
"normalized": false,
|
171 |
"special": true
|
172 |
},
|
@@ -174,8 +174,8 @@
|
|
174 |
"id": 17,
|
175 |
"content": "<unk_15>",
|
176 |
"single_word": false,
|
177 |
-
"lstrip":
|
178 |
-
"rstrip":
|
179 |
"normalized": false,
|
180 |
"special": true
|
181 |
},
|
@@ -183,8 +183,8 @@
|
|
183 |
"id": 18,
|
184 |
"content": "<unk_16>",
|
185 |
"single_word": false,
|
186 |
-
"lstrip":
|
187 |
-
"rstrip":
|
188 |
"normalized": false,
|
189 |
"special": true
|
190 |
},
|
@@ -192,8 +192,8 @@
|
|
192 |
"id": 19,
|
193 |
"content": "<unk_17>",
|
194 |
"single_word": false,
|
195 |
-
"lstrip":
|
196 |
-
"rstrip":
|
197 |
"normalized": false,
|
198 |
"special": true
|
199 |
},
|
@@ -201,8 +201,8 @@
|
|
201 |
"id": 20,
|
202 |
"content": "<unk_18>",
|
203 |
"single_word": false,
|
204 |
-
"lstrip":
|
205 |
-
"rstrip":
|
206 |
"normalized": false,
|
207 |
"special": true
|
208 |
},
|
@@ -210,8 +210,8 @@
|
|
210 |
"id": 21,
|
211 |
"content": "<unk_19>",
|
212 |
"single_word": false,
|
213 |
-
"lstrip":
|
214 |
-
"rstrip":
|
215 |
"normalized": false,
|
216 |
"special": true
|
217 |
},
|
@@ -219,8 +219,8 @@
|
|
219 |
"id": 22,
|
220 |
"content": "<unk_20>",
|
221 |
"single_word": false,
|
222 |
-
"lstrip":
|
223 |
-
"rstrip":
|
224 |
"normalized": false,
|
225 |
"special": true
|
226 |
},
|
@@ -228,8 +228,8 @@
|
|
228 |
"id": 23,
|
229 |
"content": "<unk_21>",
|
230 |
"single_word": false,
|
231 |
-
"lstrip":
|
232 |
-
"rstrip":
|
233 |
"normalized": false,
|
234 |
"special": true
|
235 |
},
|
@@ -237,8 +237,8 @@
|
|
237 |
"id": 24,
|
238 |
"content": "<unk_22>",
|
239 |
"single_word": false,
|
240 |
-
"lstrip":
|
241 |
-
"rstrip":
|
242 |
"normalized": false,
|
243 |
"special": true
|
244 |
},
|
@@ -246,8 +246,8 @@
|
|
246 |
"id": 25,
|
247 |
"content": "<unk_23>",
|
248 |
"single_word": false,
|
249 |
-
"lstrip":
|
250 |
-
"rstrip":
|
251 |
"normalized": false,
|
252 |
"special": true
|
253 |
},
|
@@ -255,8 +255,8 @@
|
|
255 |
"id": 26,
|
256 |
"content": "<unk_24>",
|
257 |
"single_word": false,
|
258 |
-
"lstrip":
|
259 |
-
"rstrip":
|
260 |
"normalized": false,
|
261 |
"special": true
|
262 |
},
|
@@ -264,8 +264,8 @@
|
|
264 |
"id": 27,
|
265 |
"content": "<unk_25>",
|
266 |
"single_word": false,
|
267 |
-
"lstrip":
|
268 |
-
"rstrip":
|
269 |
"normalized": false,
|
270 |
"special": true
|
271 |
},
|
@@ -273,8 +273,8 @@
|
|
273 |
"id": 28,
|
274 |
"content": "<unk_26>",
|
275 |
"single_word": false,
|
276 |
-
"lstrip":
|
277 |
-
"rstrip":
|
278 |
"normalized": false,
|
279 |
"special": true
|
280 |
},
|
@@ -282,8 +282,8 @@
|
|
282 |
"id": 29,
|
283 |
"content": "<unk_27>",
|
284 |
"single_word": false,
|
285 |
-
"lstrip":
|
286 |
-
"rstrip":
|
287 |
"normalized": false,
|
288 |
"special": true
|
289 |
},
|
@@ -291,8 +291,8 @@
|
|
291 |
"id": 30,
|
292 |
"content": "<unk_28>",
|
293 |
"single_word": false,
|
294 |
-
"lstrip":
|
295 |
-
"rstrip":
|
296 |
"normalized": false,
|
297 |
"special": true
|
298 |
},
|
@@ -300,8 +300,8 @@
|
|
300 |
"id": 31,
|
301 |
"content": "<unk_29>",
|
302 |
"single_word": false,
|
303 |
-
"lstrip":
|
304 |
-
"rstrip":
|
305 |
"normalized": false,
|
306 |
"special": true
|
307 |
},
|
@@ -309,8 +309,8 @@
|
|
309 |
"id": 32,
|
310 |
"content": "<unk_30>",
|
311 |
"single_word": false,
|
312 |
-
"lstrip":
|
313 |
-
"rstrip":
|
314 |
"normalized": false,
|
315 |
"special": true
|
316 |
},
|
@@ -318,8 +318,8 @@
|
|
318 |
"id": 33,
|
319 |
"content": "<unk_31>",
|
320 |
"single_word": false,
|
321 |
-
"lstrip":
|
322 |
-
"rstrip":
|
323 |
"normalized": false,
|
324 |
"special": true
|
325 |
},
|
@@ -327,8 +327,8 @@
|
|
327 |
"id": 34,
|
328 |
"content": "<unk_32>",
|
329 |
"single_word": false,
|
330 |
-
"lstrip":
|
331 |
-
"rstrip":
|
332 |
"normalized": false,
|
333 |
"special": true
|
334 |
},
|
@@ -336,8 +336,8 @@
|
|
336 |
"id": 35,
|
337 |
"content": "<unk_33>",
|
338 |
"single_word": false,
|
339 |
-
"lstrip":
|
340 |
-
"rstrip":
|
341 |
"normalized": false,
|
342 |
"special": true
|
343 |
},
|
@@ -345,8 +345,8 @@
|
|
345 |
"id": 36,
|
346 |
"content": "<unk_34>",
|
347 |
"single_word": false,
|
348 |
-
"lstrip":
|
349 |
-
"rstrip":
|
350 |
"normalized": false,
|
351 |
"special": true
|
352 |
},
|
@@ -354,8 +354,8 @@
|
|
354 |
"id": 37,
|
355 |
"content": "<unk_35>",
|
356 |
"single_word": false,
|
357 |
-
"lstrip":
|
358 |
-
"rstrip":
|
359 |
"normalized": false,
|
360 |
"special": true
|
361 |
},
|
@@ -363,8 +363,8 @@
|
|
363 |
"id": 38,
|
364 |
"content": "<unk_36>",
|
365 |
"single_word": false,
|
366 |
-
"lstrip":
|
367 |
-
"rstrip":
|
368 |
"normalized": false,
|
369 |
"special": true
|
370 |
},
|
@@ -372,8 +372,8 @@
|
|
372 |
"id": 39,
|
373 |
"content": "<unk_37>",
|
374 |
"single_word": false,
|
375 |
-
"lstrip":
|
376 |
-
"rstrip":
|
377 |
"normalized": false,
|
378 |
"special": true
|
379 |
},
|
@@ -381,8 +381,8 @@
|
|
381 |
"id": 40,
|
382 |
"content": "<unk_38>",
|
383 |
"single_word": false,
|
384 |
-
"lstrip":
|
385 |
-
"rstrip":
|
386 |
"normalized": false,
|
387 |
"special": true
|
388 |
},
|
@@ -390,8 +390,8 @@
|
|
390 |
"id": 41,
|
391 |
"content": "<unk_39>",
|
392 |
"single_word": false,
|
393 |
-
"lstrip":
|
394 |
-
"rstrip":
|
395 |
"normalized": false,
|
396 |
"special": true
|
397 |
},
|
@@ -399,8 +399,8 @@
|
|
399 |
"id": 42,
|
400 |
"content": "<unk_40>",
|
401 |
"single_word": false,
|
402 |
-
"lstrip":
|
403 |
-
"rstrip":
|
404 |
"normalized": false,
|
405 |
"special": true
|
406 |
},
|
@@ -408,8 +408,8 @@
|
|
408 |
"id": 43,
|
409 |
"content": "<unk_41>",
|
410 |
"single_word": false,
|
411 |
-
"lstrip":
|
412 |
-
"rstrip":
|
413 |
"normalized": false,
|
414 |
"special": true
|
415 |
},
|
@@ -417,8 +417,8 @@
|
|
417 |
"id": 44,
|
418 |
"content": "<unk_42>",
|
419 |
"single_word": false,
|
420 |
-
"lstrip":
|
421 |
-
"rstrip":
|
422 |
"normalized": false,
|
423 |
"special": true
|
424 |
},
|
@@ -426,8 +426,8 @@
|
|
426 |
"id": 45,
|
427 |
"content": "<unk_43>",
|
428 |
"single_word": false,
|
429 |
-
"lstrip":
|
430 |
-
"rstrip":
|
431 |
"normalized": false,
|
432 |
"special": true
|
433 |
},
|
@@ -435,8 +435,8 @@
|
|
435 |
"id": 46,
|
436 |
"content": "<unk_44>",
|
437 |
"single_word": false,
|
438 |
-
"lstrip":
|
439 |
-
"rstrip":
|
440 |
"normalized": false,
|
441 |
"special": true
|
442 |
},
|
@@ -444,8 +444,8 @@
|
|
444 |
"id": 47,
|
445 |
"content": "<unk_45>",
|
446 |
"single_word": false,
|
447 |
-
"lstrip":
|
448 |
-
"rstrip":
|
449 |
"normalized": false,
|
450 |
"special": true
|
451 |
},
|
@@ -453,8 +453,8 @@
|
|
453 |
"id": 48,
|
454 |
"content": "<unk_46>",
|
455 |
"single_word": false,
|
456 |
-
"lstrip":
|
457 |
-
"rstrip":
|
458 |
"normalized": false,
|
459 |
"special": true
|
460 |
},
|
@@ -462,8 +462,8 @@
|
|
462 |
"id": 49,
|
463 |
"content": "<unk_47>",
|
464 |
"single_word": false,
|
465 |
-
"lstrip":
|
466 |
-
"rstrip":
|
467 |
"normalized": false,
|
468 |
"special": true
|
469 |
},
|
@@ -471,8 +471,8 @@
|
|
471 |
"id": 50,
|
472 |
"content": "<unk_48>",
|
473 |
"single_word": false,
|
474 |
-
"lstrip":
|
475 |
-
"rstrip":
|
476 |
"normalized": false,
|
477 |
"special": true
|
478 |
},
|
@@ -480,8 +480,8 @@
|
|
480 |
"id": 51,
|
481 |
"content": "<unk_49>",
|
482 |
"single_word": false,
|
483 |
-
"lstrip":
|
484 |
-
"rstrip":
|
485 |
"normalized": false,
|
486 |
"special": true
|
487 |
},
|
@@ -489,8 +489,8 @@
|
|
489 |
"id": 52,
|
490 |
"content": "<unk_50>",
|
491 |
"single_word": false,
|
492 |
-
"lstrip":
|
493 |
-
"rstrip":
|
494 |
"normalized": false,
|
495 |
"special": true
|
496 |
},
|
@@ -498,8 +498,8 @@
|
|
498 |
"id": 53,
|
499 |
"content": "<unk_51>",
|
500 |
"single_word": false,
|
501 |
-
"lstrip":
|
502 |
-
"rstrip":
|
503 |
"normalized": false,
|
504 |
"special": true
|
505 |
},
|
@@ -507,8 +507,8 @@
|
|
507 |
"id": 54,
|
508 |
"content": "<unk_52>",
|
509 |
"single_word": false,
|
510 |
-
"lstrip":
|
511 |
-
"rstrip":
|
512 |
"normalized": false,
|
513 |
"special": true
|
514 |
},
|
@@ -516,8 +516,8 @@
|
|
516 |
"id": 55,
|
517 |
"content": "<unk_53>",
|
518 |
"single_word": false,
|
519 |
-
"lstrip":
|
520 |
-
"rstrip":
|
521 |
"normalized": false,
|
522 |
"special": true
|
523 |
},
|
@@ -525,8 +525,8 @@
|
|
525 |
"id": 56,
|
526 |
"content": "<unk_54>",
|
527 |
"single_word": false,
|
528 |
-
"lstrip":
|
529 |
-
"rstrip":
|
530 |
"normalized": false,
|
531 |
"special": true
|
532 |
},
|
@@ -534,8 +534,8 @@
|
|
534 |
"id": 57,
|
535 |
"content": "<unk_55>",
|
536 |
"single_word": false,
|
537 |
-
"lstrip":
|
538 |
-
"rstrip":
|
539 |
"normalized": false,
|
540 |
"special": true
|
541 |
},
|
@@ -543,8 +543,8 @@
|
|
543 |
"id": 58,
|
544 |
"content": "<unk_56>",
|
545 |
"single_word": false,
|
546 |
-
"lstrip":
|
547 |
-
"rstrip":
|
548 |
"normalized": false,
|
549 |
"special": true
|
550 |
},
|
@@ -552,8 +552,8 @@
|
|
552 |
"id": 59,
|
553 |
"content": "<unk_57>",
|
554 |
"single_word": false,
|
555 |
-
"lstrip":
|
556 |
-
"rstrip":
|
557 |
"normalized": false,
|
558 |
"special": true
|
559 |
},
|
@@ -561,8 +561,8 @@
|
|
561 |
"id": 60,
|
562 |
"content": "<unk_58>",
|
563 |
"single_word": false,
|
564 |
-
"lstrip":
|
565 |
-
"rstrip":
|
566 |
"normalized": false,
|
567 |
"special": true
|
568 |
},
|
@@ -570,8 +570,8 @@
|
|
570 |
"id": 61,
|
571 |
"content": "<unk_59>",
|
572 |
"single_word": false,
|
573 |
-
"lstrip":
|
574 |
-
"rstrip":
|
575 |
"normalized": false,
|
576 |
"special": true
|
577 |
},
|
@@ -579,8 +579,8 @@
|
|
579 |
"id": 62,
|
580 |
"content": "<unk_60>",
|
581 |
"single_word": false,
|
582 |
-
"lstrip":
|
583 |
-
"rstrip":
|
584 |
"normalized": false,
|
585 |
"special": true
|
586 |
},
|
@@ -588,8 +588,8 @@
|
|
588 |
"id": 63,
|
589 |
"content": "<unk_61>",
|
590 |
"single_word": false,
|
591 |
-
"lstrip":
|
592 |
-
"rstrip":
|
593 |
"normalized": false,
|
594 |
"special": true
|
595 |
},
|
@@ -597,8 +597,8 @@
|
|
597 |
"id": 64,
|
598 |
"content": "<unk_62>",
|
599 |
"single_word": false,
|
600 |
-
"lstrip":
|
601 |
-
"rstrip":
|
602 |
"normalized": false,
|
603 |
"special": true
|
604 |
},
|
@@ -606,8 +606,8 @@
|
|
606 |
"id": 65,
|
607 |
"content": "<unk_63>",
|
608 |
"single_word": false,
|
609 |
-
"lstrip":
|
610 |
-
"rstrip":
|
611 |
"normalized": false,
|
612 |
"special": true
|
613 |
},
|
@@ -615,8 +615,8 @@
|
|
615 |
"id": 66,
|
616 |
"content": "<unk_64>",
|
617 |
"single_word": false,
|
618 |
-
"lstrip":
|
619 |
-
"rstrip":
|
620 |
"normalized": false,
|
621 |
"special": true
|
622 |
},
|
@@ -624,8 +624,8 @@
|
|
624 |
"id": 67,
|
625 |
"content": "<unk_65>",
|
626 |
"single_word": false,
|
627 |
-
"lstrip":
|
628 |
-
"rstrip":
|
629 |
"normalized": false,
|
630 |
"special": true
|
631 |
},
|
@@ -633,8 +633,8 @@
|
|
633 |
"id": 68,
|
634 |
"content": "<unk_66>",
|
635 |
"single_word": false,
|
636 |
-
"lstrip":
|
637 |
-
"rstrip":
|
638 |
"normalized": false,
|
639 |
"special": true
|
640 |
},
|
@@ -642,8 +642,8 @@
|
|
642 |
"id": 69,
|
643 |
"content": "<unk_67>",
|
644 |
"single_word": false,
|
645 |
-
"lstrip":
|
646 |
-
"rstrip":
|
647 |
"normalized": false,
|
648 |
"special": true
|
649 |
},
|
@@ -651,8 +651,8 @@
|
|
651 |
"id": 70,
|
652 |
"content": "<unk_68>",
|
653 |
"single_word": false,
|
654 |
-
"lstrip":
|
655 |
-
"rstrip":
|
656 |
"normalized": false,
|
657 |
"special": true
|
658 |
},
|
@@ -660,8 +660,8 @@
|
|
660 |
"id": 71,
|
661 |
"content": "<unk_69>",
|
662 |
"single_word": false,
|
663 |
-
"lstrip":
|
664 |
-
"rstrip":
|
665 |
"normalized": false,
|
666 |
"special": true
|
667 |
},
|
@@ -669,8 +669,8 @@
|
|
669 |
"id": 72,
|
670 |
"content": "<unk_70>",
|
671 |
"single_word": false,
|
672 |
-
"lstrip":
|
673 |
-
"rstrip":
|
674 |
"normalized": false,
|
675 |
"special": true
|
676 |
},
|
@@ -678,8 +678,8 @@
|
|
678 |
"id": 73,
|
679 |
"content": "<unk_71>",
|
680 |
"single_word": false,
|
681 |
-
"lstrip":
|
682 |
-
"rstrip":
|
683 |
"normalized": false,
|
684 |
"special": true
|
685 |
},
|
@@ -687,8 +687,8 @@
|
|
687 |
"id": 74,
|
688 |
"content": "<unk_72>",
|
689 |
"single_word": false,
|
690 |
-
"lstrip":
|
691 |
-
"rstrip":
|
692 |
"normalized": false,
|
693 |
"special": true
|
694 |
},
|
@@ -696,8 +696,8 @@
|
|
696 |
"id": 75,
|
697 |
"content": "<unk_73>",
|
698 |
"single_word": false,
|
699 |
-
"lstrip":
|
700 |
-
"rstrip":
|
701 |
"normalized": false,
|
702 |
"special": true
|
703 |
},
|
@@ -705,8 +705,8 @@
|
|
705 |
"id": 76,
|
706 |
"content": "<unk_74>",
|
707 |
"single_word": false,
|
708 |
-
"lstrip":
|
709 |
-
"rstrip":
|
710 |
"normalized": false,
|
711 |
"special": true
|
712 |
},
|
@@ -714,8 +714,8 @@
|
|
714 |
"id": 77,
|
715 |
"content": "<unk_75>",
|
716 |
"single_word": false,
|
717 |
-
"lstrip":
|
718 |
-
"rstrip":
|
719 |
"normalized": false,
|
720 |
"special": true
|
721 |
},
|
@@ -723,8 +723,8 @@
|
|
723 |
"id": 78,
|
724 |
"content": "<unk_76>",
|
725 |
"single_word": false,
|
726 |
-
"lstrip":
|
727 |
-
"rstrip":
|
728 |
"normalized": false,
|
729 |
"special": true
|
730 |
},
|
@@ -732,8 +732,8 @@
|
|
732 |
"id": 79,
|
733 |
"content": "<unk_77>",
|
734 |
"single_word": false,
|
735 |
-
"lstrip":
|
736 |
-
"rstrip":
|
737 |
"normalized": false,
|
738 |
"special": true
|
739 |
},
|
@@ -741,8 +741,8 @@
|
|
741 |
"id": 80,
|
742 |
"content": "<unk_78>",
|
743 |
"single_word": false,
|
744 |
-
"lstrip":
|
745 |
-
"rstrip":
|
746 |
"normalized": false,
|
747 |
"special": true
|
748 |
},
|
@@ -750,8 +750,8 @@
|
|
750 |
"id": 81,
|
751 |
"content": "<unk_79>",
|
752 |
"single_word": false,
|
753 |
-
"lstrip":
|
754 |
-
"rstrip":
|
755 |
"normalized": false,
|
756 |
"special": true
|
757 |
},
|
@@ -759,8 +759,8 @@
|
|
759 |
"id": 82,
|
760 |
"content": "<unk_80>",
|
761 |
"single_word": false,
|
762 |
-
"lstrip":
|
763 |
-
"rstrip":
|
764 |
"normalized": false,
|
765 |
"special": true
|
766 |
},
|
@@ -768,8 +768,8 @@
|
|
768 |
"id": 83,
|
769 |
"content": "<unk_81>",
|
770 |
"single_word": false,
|
771 |
-
"lstrip":
|
772 |
-
"rstrip":
|
773 |
"normalized": false,
|
774 |
"special": true
|
775 |
},
|
@@ -777,8 +777,8 @@
|
|
777 |
"id": 84,
|
778 |
"content": "<unk_82>",
|
779 |
"single_word": false,
|
780 |
-
"lstrip":
|
781 |
-
"rstrip":
|
782 |
"normalized": false,
|
783 |
"special": true
|
784 |
},
|
@@ -786,8 +786,8 @@
|
|
786 |
"id": 85,
|
787 |
"content": "<unk_83>",
|
788 |
"single_word": false,
|
789 |
-
"lstrip":
|
790 |
-
"rstrip":
|
791 |
"normalized": false,
|
792 |
"special": true
|
793 |
},
|
@@ -795,8 +795,8 @@
|
|
795 |
"id": 86,
|
796 |
"content": "<unk_84>",
|
797 |
"single_word": false,
|
798 |
-
"lstrip":
|
799 |
-
"rstrip":
|
800 |
"normalized": false,
|
801 |
"special": true
|
802 |
},
|
@@ -804,8 +804,8 @@
|
|
804 |
"id": 87,
|
805 |
"content": "<unk_85>",
|
806 |
"single_word": false,
|
807 |
-
"lstrip":
|
808 |
-
"rstrip":
|
809 |
"normalized": false,
|
810 |
"special": true
|
811 |
},
|
@@ -813,8 +813,8 @@
|
|
813 |
"id": 88,
|
814 |
"content": "<unk_86>",
|
815 |
"single_word": false,
|
816 |
-
"lstrip":
|
817 |
-
"rstrip":
|
818 |
"normalized": false,
|
819 |
"special": true
|
820 |
},
|
@@ -822,8 +822,8 @@
|
|
822 |
"id": 89,
|
823 |
"content": "<unk_87>",
|
824 |
"single_word": false,
|
825 |
-
"lstrip":
|
826 |
-
"rstrip":
|
827 |
"normalized": false,
|
828 |
"special": true
|
829 |
},
|
@@ -831,8 +831,8 @@
|
|
831 |
"id": 90,
|
832 |
"content": "<unk_88>",
|
833 |
"single_word": false,
|
834 |
-
"lstrip":
|
835 |
-
"rstrip":
|
836 |
"normalized": false,
|
837 |
"special": true
|
838 |
},
|
@@ -840,8 +840,8 @@
|
|
840 |
"id": 91,
|
841 |
"content": "<unk_89>",
|
842 |
"single_word": false,
|
843 |
-
"lstrip":
|
844 |
-
"rstrip":
|
845 |
"normalized": false,
|
846 |
"special": true
|
847 |
},
|
@@ -849,8 +849,8 @@
|
|
849 |
"id": 92,
|
850 |
"content": "<unk_90>",
|
851 |
"single_word": false,
|
852 |
-
"lstrip":
|
853 |
-
"rstrip":
|
854 |
"normalized": false,
|
855 |
"special": true
|
856 |
},
|
@@ -858,8 +858,8 @@
|
|
858 |
"id": 93,
|
859 |
"content": "<unk_91>",
|
860 |
"single_word": false,
|
861 |
-
"lstrip":
|
862 |
-
"rstrip":
|
863 |
"normalized": false,
|
864 |
"special": true
|
865 |
},
|
@@ -867,8 +867,8 @@
|
|
867 |
"id": 94,
|
868 |
"content": "<unk_92>",
|
869 |
"single_word": false,
|
870 |
-
"lstrip":
|
871 |
-
"rstrip":
|
872 |
"normalized": false,
|
873 |
"special": true
|
874 |
},
|
@@ -876,8 +876,8 @@
|
|
876 |
"id": 95,
|
877 |
"content": "<unk_93>",
|
878 |
"single_word": false,
|
879 |
-
"lstrip":
|
880 |
-
"rstrip":
|
881 |
"normalized": false,
|
882 |
"special": true
|
883 |
},
|
@@ -885,8 +885,8 @@
|
|
885 |
"id": 96,
|
886 |
"content": "<unk_94>",
|
887 |
"single_word": false,
|
888 |
-
"lstrip":
|
889 |
-
"rstrip":
|
890 |
"normalized": false,
|
891 |
"special": true
|
892 |
},
|
@@ -894,8 +894,8 @@
|
|
894 |
"id": 97,
|
895 |
"content": "<unk_95>",
|
896 |
"single_word": false,
|
897 |
-
"lstrip":
|
898 |
-
"rstrip":
|
899 |
"normalized": false,
|
900 |
"special": true
|
901 |
},
|
@@ -903,8 +903,8 @@
|
|
903 |
"id": 98,
|
904 |
"content": "<unk_96>",
|
905 |
"single_word": false,
|
906 |
-
"lstrip":
|
907 |
-
"rstrip":
|
908 |
"normalized": false,
|
909 |
"special": true
|
910 |
},
|
@@ -912,8 +912,8 @@
|
|
912 |
"id": 99,
|
913 |
"content": "<unk_97>",
|
914 |
"single_word": false,
|
915 |
-
"lstrip":
|
916 |
-
"rstrip":
|
917 |
"normalized": false,
|
918 |
"special": true
|
919 |
},
|
@@ -921,8 +921,8 @@
|
|
921 |
"id": 100,
|
922 |
"content": "<unk_98>",
|
923 |
"single_word": false,
|
924 |
-
"lstrip":
|
925 |
-
"rstrip":
|
926 |
"normalized": false,
|
927 |
"special": true
|
928 |
},
|
@@ -930,8 +930,8 @@
|
|
930 |
"id": 101,
|
931 |
"content": "<unk_99>",
|
932 |
"single_word": false,
|
933 |
-
"lstrip":
|
934 |
-
"rstrip":
|
935 |
"normalized": false,
|
936 |
"special": true
|
937 |
},
|
@@ -939,8 +939,8 @@
|
|
939 |
"id": 102,
|
940 |
"content": "<unk_100>",
|
941 |
"single_word": false,
|
942 |
-
"lstrip":
|
943 |
-
"rstrip":
|
944 |
"normalized": false,
|
945 |
"special": true
|
946 |
},
|
@@ -948,8 +948,8 @@
|
|
948 |
"id": 103,
|
949 |
"content": "<unk_101>",
|
950 |
"single_word": false,
|
951 |
-
"lstrip":
|
952 |
-
"rstrip":
|
953 |
"normalized": false,
|
954 |
"special": true
|
955 |
},
|
@@ -957,8 +957,8 @@
|
|
957 |
"id": 104,
|
958 |
"content": "<unk_102>",
|
959 |
"single_word": false,
|
960 |
-
"lstrip":
|
961 |
-
"rstrip":
|
962 |
"normalized": false,
|
963 |
"special": true
|
964 |
},
|
|
|
39 |
"id": 2,
|
40 |
"content": "<mask_1>",
|
41 |
"single_word": false,
|
42 |
+
"lstrip": false,
|
43 |
+
"rstrip": false,
|
44 |
"normalized": false,
|
45 |
"special": true
|
46 |
},
|
|
|
57 |
"id": 4,
|
58 |
"content": "<unk_2>",
|
59 |
"single_word": false,
|
60 |
+
"lstrip": false,
|
61 |
+
"rstrip": false,
|
62 |
"normalized": false,
|
63 |
"special": true
|
64 |
},
|
|
|
66 |
"id": 5,
|
67 |
"content": "<unk_3>",
|
68 |
"single_word": false,
|
69 |
+
"lstrip": false,
|
70 |
+
"rstrip": false,
|
71 |
"normalized": false,
|
72 |
"special": true
|
73 |
},
|
|
|
75 |
"id": 6,
|
76 |
"content": "<unk_4>",
|
77 |
"single_word": false,
|
78 |
+
"lstrip": false,
|
79 |
+
"rstrip": false,
|
80 |
"normalized": false,
|
81 |
"special": true
|
82 |
},
|
|
|
84 |
"id": 7,
|
85 |
"content": "<unk_5>",
|
86 |
"single_word": false,
|
87 |
+
"lstrip": false,
|
88 |
+
"rstrip": false,
|
89 |
"normalized": false,
|
90 |
"special": true
|
91 |
},
|
|
|
93 |
"id": 8,
|
94 |
"content": "<unk_6>",
|
95 |
"single_word": false,
|
96 |
+
"lstrip": false,
|
97 |
+
"rstrip": false,
|
98 |
"normalized": false,
|
99 |
"special": true
|
100 |
},
|
|
|
102 |
"id": 9,
|
103 |
"content": "<unk_7>",
|
104 |
"single_word": false,
|
105 |
+
"lstrip": false,
|
106 |
+
"rstrip": false,
|
107 |
"normalized": false,
|
108 |
"special": true
|
109 |
},
|
|
|
111 |
"id": 10,
|
112 |
"content": "<unk_8>",
|
113 |
"single_word": false,
|
114 |
+
"lstrip": false,
|
115 |
+
"rstrip": false,
|
116 |
"normalized": false,
|
117 |
"special": true
|
118 |
},
|
|
|
120 |
"id": 11,
|
121 |
"content": "<unk_9>",
|
122 |
"single_word": false,
|
123 |
+
"lstrip": false,
|
124 |
+
"rstrip": false,
|
125 |
"normalized": false,
|
126 |
"special": true
|
127 |
},
|
|
|
129 |
"id": 12,
|
130 |
"content": "<unk_10>",
|
131 |
"single_word": false,
|
132 |
+
"lstrip": false,
|
133 |
+
"rstrip": false,
|
134 |
"normalized": false,
|
135 |
"special": true
|
136 |
},
|
|
|
138 |
"id": 13,
|
139 |
"content": "<unk_11>",
|
140 |
"single_word": false,
|
141 |
+
"lstrip": false,
|
142 |
+
"rstrip": false,
|
143 |
"normalized": false,
|
144 |
"special": true
|
145 |
},
|
|
|
147 |
"id": 14,
|
148 |
"content": "<unk_12>",
|
149 |
"single_word": false,
|
150 |
+
"lstrip": false,
|
151 |
+
"rstrip": false,
|
152 |
"normalized": false,
|
153 |
"special": true
|
154 |
},
|
|
|
156 |
"id": 15,
|
157 |
"content": "<unk_13>",
|
158 |
"single_word": false,
|
159 |
+
"lstrip": false,
|
160 |
+
"rstrip": false,
|
161 |
"normalized": false,
|
162 |
"special": true
|
163 |
},
|
|
|
165 |
"id": 16,
|
166 |
"content": "<unk_14>",
|
167 |
"single_word": false,
|
168 |
+
"lstrip": false,
|
169 |
+
"rstrip": false,
|
170 |
"normalized": false,
|
171 |
"special": true
|
172 |
},
|
|
|
174 |
"id": 17,
|
175 |
"content": "<unk_15>",
|
176 |
"single_word": false,
|
177 |
+
"lstrip": false,
|
178 |
+
"rstrip": false,
|
179 |
"normalized": false,
|
180 |
"special": true
|
181 |
},
|
|
|
183 |
"id": 18,
|
184 |
"content": "<unk_16>",
|
185 |
"single_word": false,
|
186 |
+
"lstrip": false,
|
187 |
+
"rstrip": false,
|
188 |
"normalized": false,
|
189 |
"special": true
|
190 |
},
|
|
|
192 |
"id": 19,
|
193 |
"content": "<unk_17>",
|
194 |
"single_word": false,
|
195 |
+
"lstrip": false,
|
196 |
+
"rstrip": false,
|
197 |
"normalized": false,
|
198 |
"special": true
|
199 |
},
|
|
|
201 |
"id": 20,
|
202 |
"content": "<unk_18>",
|
203 |
"single_word": false,
|
204 |
+
"lstrip": false,
|
205 |
+
"rstrip": false,
|
206 |
"normalized": false,
|
207 |
"special": true
|
208 |
},
|
|
|
210 |
"id": 21,
|
211 |
"content": "<unk_19>",
|
212 |
"single_word": false,
|
213 |
+
"lstrip": false,
|
214 |
+
"rstrip": false,
|
215 |
"normalized": false,
|
216 |
"special": true
|
217 |
},
|
|
|
219 |
"id": 22,
|
220 |
"content": "<unk_20>",
|
221 |
"single_word": false,
|
222 |
+
"lstrip": false,
|
223 |
+
"rstrip": false,
|
224 |
"normalized": false,
|
225 |
"special": true
|
226 |
},
|
|
|
228 |
"id": 23,
|
229 |
"content": "<unk_21>",
|
230 |
"single_word": false,
|
231 |
+
"lstrip": false,
|
232 |
+
"rstrip": false,
|
233 |
"normalized": false,
|
234 |
"special": true
|
235 |
},
|
|
|
237 |
"id": 24,
|
238 |
"content": "<unk_22>",
|
239 |
"single_word": false,
|
240 |
+
"lstrip": false,
|
241 |
+
"rstrip": false,
|
242 |
"normalized": false,
|
243 |
"special": true
|
244 |
},
|
|
|
246 |
"id": 25,
|
247 |
"content": "<unk_23>",
|
248 |
"single_word": false,
|
249 |
+
"lstrip": false,
|
250 |
+
"rstrip": false,
|
251 |
"normalized": false,
|
252 |
"special": true
|
253 |
},
|
|
|
255 |
"id": 26,
|
256 |
"content": "<unk_24>",
|
257 |
"single_word": false,
|
258 |
+
"lstrip": false,
|
259 |
+
"rstrip": false,
|
260 |
"normalized": false,
|
261 |
"special": true
|
262 |
},
|
|
|
264 |
"id": 27,
|
265 |
"content": "<unk_25>",
|
266 |
"single_word": false,
|
267 |
+
"lstrip": false,
|
268 |
+
"rstrip": false,
|
269 |
"normalized": false,
|
270 |
"special": true
|
271 |
},
|
|
|
273 |
"id": 28,
|
274 |
"content": "<unk_26>",
|
275 |
"single_word": false,
|
276 |
+
"lstrip": false,
|
277 |
+
"rstrip": false,
|
278 |
"normalized": false,
|
279 |
"special": true
|
280 |
},
|
|
|
282 |
"id": 29,
|
283 |
"content": "<unk_27>",
|
284 |
"single_word": false,
|
285 |
+
"lstrip": false,
|
286 |
+
"rstrip": false,
|
287 |
"normalized": false,
|
288 |
"special": true
|
289 |
},
|
|
|
291 |
"id": 30,
|
292 |
"content": "<unk_28>",
|
293 |
"single_word": false,
|
294 |
+
"lstrip": false,
|
295 |
+
"rstrip": false,
|
296 |
"normalized": false,
|
297 |
"special": true
|
298 |
},
|
|
|
300 |
"id": 31,
|
301 |
"content": "<unk_29>",
|
302 |
"single_word": false,
|
303 |
+
"lstrip": false,
|
304 |
+
"rstrip": false,
|
305 |
"normalized": false,
|
306 |
"special": true
|
307 |
},
|
|
|
309 |
"id": 32,
|
310 |
"content": "<unk_30>",
|
311 |
"single_word": false,
|
312 |
+
"lstrip": false,
|
313 |
+
"rstrip": false,
|
314 |
"normalized": false,
|
315 |
"special": true
|
316 |
},
|
|
|
318 |
"id": 33,
|
319 |
"content": "<unk_31>",
|
320 |
"single_word": false,
|
321 |
+
"lstrip": false,
|
322 |
+
"rstrip": false,
|
323 |
"normalized": false,
|
324 |
"special": true
|
325 |
},
|
|
|
327 |
"id": 34,
|
328 |
"content": "<unk_32>",
|
329 |
"single_word": false,
|
330 |
+
"lstrip": false,
|
331 |
+
"rstrip": false,
|
332 |
"normalized": false,
|
333 |
"special": true
|
334 |
},
|
|
|
336 |
"id": 35,
|
337 |
"content": "<unk_33>",
|
338 |
"single_word": false,
|
339 |
+
"lstrip": false,
|
340 |
+
"rstrip": false,
|
341 |
"normalized": false,
|
342 |
"special": true
|
343 |
},
|
|
|
345 |
"id": 36,
|
346 |
"content": "<unk_34>",
|
347 |
"single_word": false,
|
348 |
+
"lstrip": false,
|
349 |
+
"rstrip": false,
|
350 |
"normalized": false,
|
351 |
"special": true
|
352 |
},
|
|
|
354 |
"id": 37,
|
355 |
"content": "<unk_35>",
|
356 |
"single_word": false,
|
357 |
+
"lstrip": false,
|
358 |
+
"rstrip": false,
|
359 |
"normalized": false,
|
360 |
"special": true
|
361 |
},
|
|
|
363 |
"id": 38,
|
364 |
"content": "<unk_36>",
|
365 |
"single_word": false,
|
366 |
+
"lstrip": false,
|
367 |
+
"rstrip": false,
|
368 |
"normalized": false,
|
369 |
"special": true
|
370 |
},
|
|
|
372 |
"id": 39,
|
373 |
"content": "<unk_37>",
|
374 |
"single_word": false,
|
375 |
+
"lstrip": false,
|
376 |
+
"rstrip": false,
|
377 |
"normalized": false,
|
378 |
"special": true
|
379 |
},
|
|
|
381 |
"id": 40,
|
382 |
"content": "<unk_38>",
|
383 |
"single_word": false,
|
384 |
+
"lstrip": false,
|
385 |
+
"rstrip": false,
|
386 |
"normalized": false,
|
387 |
"special": true
|
388 |
},
|
|
|
390 |
"id": 41,
|
391 |
"content": "<unk_39>",
|
392 |
"single_word": false,
|
393 |
+
"lstrip": false,
|
394 |
+
"rstrip": false,
|
395 |
"normalized": false,
|
396 |
"special": true
|
397 |
},
|
|
|
399 |
"id": 42,
|
400 |
"content": "<unk_40>",
|
401 |
"single_word": false,
|
402 |
+
"lstrip": false,
|
403 |
+
"rstrip": false,
|
404 |
"normalized": false,
|
405 |
"special": true
|
406 |
},
|
|
|
408 |
"id": 43,
|
409 |
"content": "<unk_41>",
|
410 |
"single_word": false,
|
411 |
+
"lstrip": false,
|
412 |
+
"rstrip": false,
|
413 |
"normalized": false,
|
414 |
"special": true
|
415 |
},
|
|
|
417 |
"id": 44,
|
418 |
"content": "<unk_42>",
|
419 |
"single_word": false,
|
420 |
+
"lstrip": false,
|
421 |
+
"rstrip": false,
|
422 |
"normalized": false,
|
423 |
"special": true
|
424 |
},
|
|
|
426 |
"id": 45,
|
427 |
"content": "<unk_43>",
|
428 |
"single_word": false,
|
429 |
+
"lstrip": false,
|
430 |
+
"rstrip": false,
|
431 |
"normalized": false,
|
432 |
"special": true
|
433 |
},
|
|
|
435 |
"id": 46,
|
436 |
"content": "<unk_44>",
|
437 |
"single_word": false,
|
438 |
+
"lstrip": false,
|
439 |
+
"rstrip": false,
|
440 |
"normalized": false,
|
441 |
"special": true
|
442 |
},
|
|
|
444 |
"id": 47,
|
445 |
"content": "<unk_45>",
|
446 |
"single_word": false,
|
447 |
+
"lstrip": false,
|
448 |
+
"rstrip": false,
|
449 |
"normalized": false,
|
450 |
"special": true
|
451 |
},
|
|
|
453 |
"id": 48,
|
454 |
"content": "<unk_46>",
|
455 |
"single_word": false,
|
456 |
+
"lstrip": false,
|
457 |
+
"rstrip": false,
|
458 |
"normalized": false,
|
459 |
"special": true
|
460 |
},
|
|
|
462 |
"id": 49,
|
463 |
"content": "<unk_47>",
|
464 |
"single_word": false,
|
465 |
+
"lstrip": false,
|
466 |
+
"rstrip": false,
|
467 |
"normalized": false,
|
468 |
"special": true
|
469 |
},
|
|
|
471 |
"id": 50,
|
472 |
"content": "<unk_48>",
|
473 |
"single_word": false,
|
474 |
+
"lstrip": false,
|
475 |
+
"rstrip": false,
|
476 |
"normalized": false,
|
477 |
"special": true
|
478 |
},
|
|
|
480 |
"id": 51,
|
481 |
"content": "<unk_49>",
|
482 |
"single_word": false,
|
483 |
+
"lstrip": false,
|
484 |
+
"rstrip": false,
|
485 |
"normalized": false,
|
486 |
"special": true
|
487 |
},
|
|
|
489 |
"id": 52,
|
490 |
"content": "<unk_50>",
|
491 |
"single_word": false,
|
492 |
+
"lstrip": false,
|
493 |
+
"rstrip": false,
|
494 |
"normalized": false,
|
495 |
"special": true
|
496 |
},
|
|
|
498 |
"id": 53,
|
499 |
"content": "<unk_51>",
|
500 |
"single_word": false,
|
501 |
+
"lstrip": false,
|
502 |
+
"rstrip": false,
|
503 |
"normalized": false,
|
504 |
"special": true
|
505 |
},
|
|
|
507 |
"id": 54,
|
508 |
"content": "<unk_52>",
|
509 |
"single_word": false,
|
510 |
+
"lstrip": false,
|
511 |
+
"rstrip": false,
|
512 |
"normalized": false,
|
513 |
"special": true
|
514 |
},
|
|
|
516 |
"id": 55,
|
517 |
"content": "<unk_53>",
|
518 |
"single_word": false,
|
519 |
+
"lstrip": false,
|
520 |
+
"rstrip": false,
|
521 |
"normalized": false,
|
522 |
"special": true
|
523 |
},
|
|
|
525 |
"id": 56,
|
526 |
"content": "<unk_54>",
|
527 |
"single_word": false,
|
528 |
+
"lstrip": false,
|
529 |
+
"rstrip": false,
|
530 |
"normalized": false,
|
531 |
"special": true
|
532 |
},
|
|
|
534 |
"id": 57,
|
535 |
"content": "<unk_55>",
|
536 |
"single_word": false,
|
537 |
+
"lstrip": false,
|
538 |
+
"rstrip": false,
|
539 |
"normalized": false,
|
540 |
"special": true
|
541 |
},
|
|
|
543 |
"id": 58,
|
544 |
"content": "<unk_56>",
|
545 |
"single_word": false,
|
546 |
+
"lstrip": false,
|
547 |
+
"rstrip": false,
|
548 |
"normalized": false,
|
549 |
"special": true
|
550 |
},
|
|
|
552 |
"id": 59,
|
553 |
"content": "<unk_57>",
|
554 |
"single_word": false,
|
555 |
+
"lstrip": false,
|
556 |
+
"rstrip": false,
|
557 |
"normalized": false,
|
558 |
"special": true
|
559 |
},
|
|
|
561 |
"id": 60,
|
562 |
"content": "<unk_58>",
|
563 |
"single_word": false,
|
564 |
+
"lstrip": false,
|
565 |
+
"rstrip": false,
|
566 |
"normalized": false,
|
567 |
"special": true
|
568 |
},
|
|
|
570 |
"id": 61,
|
571 |
"content": "<unk_59>",
|
572 |
"single_word": false,
|
573 |
+
"lstrip": false,
|
574 |
+
"rstrip": false,
|
575 |
"normalized": false,
|
576 |
"special": true
|
577 |
},
|
|
|
579 |
"id": 62,
|
580 |
"content": "<unk_60>",
|
581 |
"single_word": false,
|
582 |
+
"lstrip": false,
|
583 |
+
"rstrip": false,
|
584 |
"normalized": false,
|
585 |
"special": true
|
586 |
},
|
|
|
588 |
"id": 63,
|
589 |
"content": "<unk_61>",
|
590 |
"single_word": false,
|
591 |
+
"lstrip": false,
|
592 |
+
"rstrip": false,
|
593 |
"normalized": false,
|
594 |
"special": true
|
595 |
},
|
|
|
597 |
"id": 64,
|
598 |
"content": "<unk_62>",
|
599 |
"single_word": false,
|
600 |
+
"lstrip": false,
|
601 |
+
"rstrip": false,
|
602 |
"normalized": false,
|
603 |
"special": true
|
604 |
},
|
|
|
606 |
"id": 65,
|
607 |
"content": "<unk_63>",
|
608 |
"single_word": false,
|
609 |
+
"lstrip": false,
|
610 |
+
"rstrip": false,
|
611 |
"normalized": false,
|
612 |
"special": true
|
613 |
},
|
|
|
615 |
"id": 66,
|
616 |
"content": "<unk_64>",
|
617 |
"single_word": false,
|
618 |
+
"lstrip": false,
|
619 |
+
"rstrip": false,
|
620 |
"normalized": false,
|
621 |
"special": true
|
622 |
},
|
|
|
624 |
"id": 67,
|
625 |
"content": "<unk_65>",
|
626 |
"single_word": false,
|
627 |
+
"lstrip": false,
|
628 |
+
"rstrip": false,
|
629 |
"normalized": false,
|
630 |
"special": true
|
631 |
},
|
|
|
633 |
"id": 68,
|
634 |
"content": "<unk_66>",
|
635 |
"single_word": false,
|
636 |
+
"lstrip": false,
|
637 |
+
"rstrip": false,
|
638 |
"normalized": false,
|
639 |
"special": true
|
640 |
},
|
|
|
642 |
"id": 69,
|
643 |
"content": "<unk_67>",
|
644 |
"single_word": false,
|
645 |
+
"lstrip": false,
|
646 |
+
"rstrip": false,
|
647 |
"normalized": false,
|
648 |
"special": true
|
649 |
},
|
|
|
651 |
"id": 70,
|
652 |
"content": "<unk_68>",
|
653 |
"single_word": false,
|
654 |
+
"lstrip": false,
|
655 |
+
"rstrip": false,
|
656 |
"normalized": false,
|
657 |
"special": true
|
658 |
},
|
|
|
660 |
"id": 71,
|
661 |
"content": "<unk_69>",
|
662 |
"single_word": false,
|
663 |
+
"lstrip": false,
|
664 |
+
"rstrip": false,
|
665 |
"normalized": false,
|
666 |
"special": true
|
667 |
},
|
|
|
669 |
"id": 72,
|
670 |
"content": "<unk_70>",
|
671 |
"single_word": false,
|
672 |
+
"lstrip": false,
|
673 |
+
"rstrip": false,
|
674 |
"normalized": false,
|
675 |
"special": true
|
676 |
},
|
|
|
678 |
"id": 73,
|
679 |
"content": "<unk_71>",
|
680 |
"single_word": false,
|
681 |
+
"lstrip": false,
|
682 |
+
"rstrip": false,
|
683 |
"normalized": false,
|
684 |
"special": true
|
685 |
},
|
|
|
687 |
"id": 74,
|
688 |
"content": "<unk_72>",
|
689 |
"single_word": false,
|
690 |
+
"lstrip": false,
|
691 |
+
"rstrip": false,
|
692 |
"normalized": false,
|
693 |
"special": true
|
694 |
},
|
|
|
696 |
"id": 75,
|
697 |
"content": "<unk_73>",
|
698 |
"single_word": false,
|
699 |
+
"lstrip": false,
|
700 |
+
"rstrip": false,
|
701 |
"normalized": false,
|
702 |
"special": true
|
703 |
},
|
|
|
705 |
"id": 76,
|
706 |
"content": "<unk_74>",
|
707 |
"single_word": false,
|
708 |
+
"lstrip": false,
|
709 |
+
"rstrip": false,
|
710 |
"normalized": false,
|
711 |
"special": true
|
712 |
},
|
|
|
714 |
"id": 77,
|
715 |
"content": "<unk_75>",
|
716 |
"single_word": false,
|
717 |
+
"lstrip": false,
|
718 |
+
"rstrip": false,
|
719 |
"normalized": false,
|
720 |
"special": true
|
721 |
},
|
|
|
723 |
"id": 78,
|
724 |
"content": "<unk_76>",
|
725 |
"single_word": false,
|
726 |
+
"lstrip": false,
|
727 |
+
"rstrip": false,
|
728 |
"normalized": false,
|
729 |
"special": true
|
730 |
},
|
|
|
732 |
"id": 79,
|
733 |
"content": "<unk_77>",
|
734 |
"single_word": false,
|
735 |
+
"lstrip": false,
|
736 |
+
"rstrip": false,
|
737 |
"normalized": false,
|
738 |
"special": true
|
739 |
},
|
|
|
741 |
"id": 80,
|
742 |
"content": "<unk_78>",
|
743 |
"single_word": false,
|
744 |
+
"lstrip": false,
|
745 |
+
"rstrip": false,
|
746 |
"normalized": false,
|
747 |
"special": true
|
748 |
},
|
|
|
750 |
"id": 81,
|
751 |
"content": "<unk_79>",
|
752 |
"single_word": false,
|
753 |
+
"lstrip": false,
|
754 |
+
"rstrip": false,
|
755 |
"normalized": false,
|
756 |
"special": true
|
757 |
},
|
|
|
759 |
"id": 82,
|
760 |
"content": "<unk_80>",
|
761 |
"single_word": false,
|
762 |
+
"lstrip": false,
|
763 |
+
"rstrip": false,
|
764 |
"normalized": false,
|
765 |
"special": true
|
766 |
},
|
|
|
768 |
"id": 83,
|
769 |
"content": "<unk_81>",
|
770 |
"single_word": false,
|
771 |
+
"lstrip": false,
|
772 |
+
"rstrip": false,
|
773 |
"normalized": false,
|
774 |
"special": true
|
775 |
},
|
|
|
777 |
"id": 84,
|
778 |
"content": "<unk_82>",
|
779 |
"single_word": false,
|
780 |
+
"lstrip": false,
|
781 |
+
"rstrip": false,
|
782 |
"normalized": false,
|
783 |
"special": true
|
784 |
},
|
|
|
786 |
"id": 85,
|
787 |
"content": "<unk_83>",
|
788 |
"single_word": false,
|
789 |
+
"lstrip": false,
|
790 |
+
"rstrip": false,
|
791 |
"normalized": false,
|
792 |
"special": true
|
793 |
},
|
|
|
795 |
"id": 86,
|
796 |
"content": "<unk_84>",
|
797 |
"single_word": false,
|
798 |
+
"lstrip": false,
|
799 |
+
"rstrip": false,
|
800 |
"normalized": false,
|
801 |
"special": true
|
802 |
},
|
|
|
804 |
"id": 87,
|
805 |
"content": "<unk_85>",
|
806 |
"single_word": false,
|
807 |
+
"lstrip": false,
|
808 |
+
"rstrip": false,
|
809 |
"normalized": false,
|
810 |
"special": true
|
811 |
},
|
|
|
813 |
"id": 88,
|
814 |
"content": "<unk_86>",
|
815 |
"single_word": false,
|
816 |
+
"lstrip": false,
|
817 |
+
"rstrip": false,
|
818 |
"normalized": false,
|
819 |
"special": true
|
820 |
},
|
|
|
822 |
"id": 89,
|
823 |
"content": "<unk_87>",
|
824 |
"single_word": false,
|
825 |
+
"lstrip": false,
|
826 |
+
"rstrip": false,
|
827 |
"normalized": false,
|
828 |
"special": true
|
829 |
},
|
|
|
831 |
"id": 90,
|
832 |
"content": "<unk_88>",
|
833 |
"single_word": false,
|
834 |
+
"lstrip": false,
|
835 |
+
"rstrip": false,
|
836 |
"normalized": false,
|
837 |
"special": true
|
838 |
},
|
|
|
840 |
"id": 91,
|
841 |
"content": "<unk_89>",
|
842 |
"single_word": false,
|
843 |
+
"lstrip": false,
|
844 |
+
"rstrip": false,
|
845 |
"normalized": false,
|
846 |
"special": true
|
847 |
},
|
|
|
849 |
"id": 92,
|
850 |
"content": "<unk_90>",
|
851 |
"single_word": false,
|
852 |
+
"lstrip": false,
|
853 |
+
"rstrip": false,
|
854 |
"normalized": false,
|
855 |
"special": true
|
856 |
},
|
|
|
858 |
"id": 93,
|
859 |
"content": "<unk_91>",
|
860 |
"single_word": false,
|
861 |
+
"lstrip": false,
|
862 |
+
"rstrip": false,
|
863 |
"normalized": false,
|
864 |
"special": true
|
865 |
},
|
|
|
867 |
"id": 94,
|
868 |
"content": "<unk_92>",
|
869 |
"single_word": false,
|
870 |
+
"lstrip": false,
|
871 |
+
"rstrip": false,
|
872 |
"normalized": false,
|
873 |
"special": true
|
874 |
},
|
|
|
876 |
"id": 95,
|
877 |
"content": "<unk_93>",
|
878 |
"single_word": false,
|
879 |
+
"lstrip": false,
|
880 |
+
"rstrip": false,
|
881 |
"normalized": false,
|
882 |
"special": true
|
883 |
},
|
|
|
885 |
"id": 96,
|
886 |
"content": "<unk_94>",
|
887 |
"single_word": false,
|
888 |
+
"lstrip": false,
|
889 |
+
"rstrip": false,
|
890 |
"normalized": false,
|
891 |
"special": true
|
892 |
},
|
|
|
894 |
"id": 97,
|
895 |
"content": "<unk_95>",
|
896 |
"single_word": false,
|
897 |
+
"lstrip": false,
|
898 |
+
"rstrip": false,
|
899 |
"normalized": false,
|
900 |
"special": true
|
901 |
},
|
|
|
903 |
"id": 98,
|
904 |
"content": "<unk_96>",
|
905 |
"single_word": false,
|
906 |
+
"lstrip": false,
|
907 |
+
"rstrip": false,
|
908 |
"normalized": false,
|
909 |
"special": true
|
910 |
},
|
|
|
912 |
"id": 99,
|
913 |
"content": "<unk_97>",
|
914 |
"single_word": false,
|
915 |
+
"lstrip": false,
|
916 |
+
"rstrip": false,
|
917 |
"normalized": false,
|
918 |
"special": true
|
919 |
},
|
|
|
921 |
"id": 100,
|
922 |
"content": "<unk_98>",
|
923 |
"single_word": false,
|
924 |
+
"lstrip": false,
|
925 |
+
"rstrip": false,
|
926 |
"normalized": false,
|
927 |
"special": true
|
928 |
},
|
|
|
930 |
"id": 101,
|
931 |
"content": "<unk_99>",
|
932 |
"single_word": false,
|
933 |
+
"lstrip": false,
|
934 |
+
"rstrip": false,
|
935 |
"normalized": false,
|
936 |
"special": true
|
937 |
},
|
|
|
939 |
"id": 102,
|
940 |
"content": "<unk_100>",
|
941 |
"single_word": false,
|
942 |
+
"lstrip": false,
|
943 |
+
"rstrip": false,
|
944 |
"normalized": false,
|
945 |
"special": true
|
946 |
},
|
|
|
948 |
"id": 103,
|
949 |
"content": "<unk_101>",
|
950 |
"single_word": false,
|
951 |
+
"lstrip": false,
|
952 |
+
"rstrip": false,
|
953 |
"normalized": false,
|
954 |
"special": true
|
955 |
},
|
|
|
957 |
"id": 104,
|
958 |
"content": "<unk_102>",
|
959 |
"single_word": false,
|
960 |
+
"lstrip": false,
|
961 |
+
"rstrip": false,
|
962 |
"normalized": false,
|
963 |
"special": true
|
964 |
},
|
tokenizer_config.json
CHANGED
@@ -18,9 +18,9 @@
|
|
18 |
},
|
19 |
"2": {
|
20 |
"content": "<mask_1>",
|
21 |
-
"lstrip":
|
22 |
"normalized": false,
|
23 |
-
"rstrip":
|
24 |
"single_word": false,
|
25 |
"special": true
|
26 |
},
|
@@ -34,809 +34,809 @@
|
|
34 |
},
|
35 |
"4": {
|
36 |
"content": "<unk_2>",
|
37 |
-
"lstrip":
|
38 |
"normalized": false,
|
39 |
-
"rstrip":
|
40 |
"single_word": false,
|
41 |
"special": true
|
42 |
},
|
43 |
"5": {
|
44 |
"content": "<unk_3>",
|
45 |
-
"lstrip":
|
46 |
"normalized": false,
|
47 |
-
"rstrip":
|
48 |
"single_word": false,
|
49 |
"special": true
|
50 |
},
|
51 |
"6": {
|
52 |
"content": "<unk_4>",
|
53 |
-
"lstrip":
|
54 |
"normalized": false,
|
55 |
-
"rstrip":
|
56 |
"single_word": false,
|
57 |
"special": true
|
58 |
},
|
59 |
"7": {
|
60 |
"content": "<unk_5>",
|
61 |
-
"lstrip":
|
62 |
"normalized": false,
|
63 |
-
"rstrip":
|
64 |
"single_word": false,
|
65 |
"special": true
|
66 |
},
|
67 |
"8": {
|
68 |
"content": "<unk_6>",
|
69 |
-
"lstrip":
|
70 |
"normalized": false,
|
71 |
-
"rstrip":
|
72 |
"single_word": false,
|
73 |
"special": true
|
74 |
},
|
75 |
"9": {
|
76 |
"content": "<unk_7>",
|
77 |
-
"lstrip":
|
78 |
"normalized": false,
|
79 |
-
"rstrip":
|
80 |
"single_word": false,
|
81 |
"special": true
|
82 |
},
|
83 |
"10": {
|
84 |
"content": "<unk_8>",
|
85 |
-
"lstrip":
|
86 |
"normalized": false,
|
87 |
-
"rstrip":
|
88 |
"single_word": false,
|
89 |
"special": true
|
90 |
},
|
91 |
"11": {
|
92 |
"content": "<unk_9>",
|
93 |
-
"lstrip":
|
94 |
"normalized": false,
|
95 |
-
"rstrip":
|
96 |
"single_word": false,
|
97 |
"special": true
|
98 |
},
|
99 |
"12": {
|
100 |
"content": "<unk_10>",
|
101 |
-
"lstrip":
|
102 |
"normalized": false,
|
103 |
-
"rstrip":
|
104 |
"single_word": false,
|
105 |
"special": true
|
106 |
},
|
107 |
"13": {
|
108 |
"content": "<unk_11>",
|
109 |
-
"lstrip":
|
110 |
"normalized": false,
|
111 |
-
"rstrip":
|
112 |
"single_word": false,
|
113 |
"special": true
|
114 |
},
|
115 |
"14": {
|
116 |
"content": "<unk_12>",
|
117 |
-
"lstrip":
|
118 |
"normalized": false,
|
119 |
-
"rstrip":
|
120 |
"single_word": false,
|
121 |
"special": true
|
122 |
},
|
123 |
"15": {
|
124 |
"content": "<unk_13>",
|
125 |
-
"lstrip":
|
126 |
"normalized": false,
|
127 |
-
"rstrip":
|
128 |
"single_word": false,
|
129 |
"special": true
|
130 |
},
|
131 |
"16": {
|
132 |
"content": "<unk_14>",
|
133 |
-
"lstrip":
|
134 |
"normalized": false,
|
135 |
-
"rstrip":
|
136 |
"single_word": false,
|
137 |
"special": true
|
138 |
},
|
139 |
"17": {
|
140 |
"content": "<unk_15>",
|
141 |
-
"lstrip":
|
142 |
"normalized": false,
|
143 |
-
"rstrip":
|
144 |
"single_word": false,
|
145 |
"special": true
|
146 |
},
|
147 |
"18": {
|
148 |
"content": "<unk_16>",
|
149 |
-
"lstrip":
|
150 |
"normalized": false,
|
151 |
-
"rstrip":
|
152 |
"single_word": false,
|
153 |
"special": true
|
154 |
},
|
155 |
"19": {
|
156 |
"content": "<unk_17>",
|
157 |
-
"lstrip":
|
158 |
"normalized": false,
|
159 |
-
"rstrip":
|
160 |
"single_word": false,
|
161 |
"special": true
|
162 |
},
|
163 |
"20": {
|
164 |
"content": "<unk_18>",
|
165 |
-
"lstrip":
|
166 |
"normalized": false,
|
167 |
-
"rstrip":
|
168 |
"single_word": false,
|
169 |
"special": true
|
170 |
},
|
171 |
"21": {
|
172 |
"content": "<unk_19>",
|
173 |
-
"lstrip":
|
174 |
"normalized": false,
|
175 |
-
"rstrip":
|
176 |
"single_word": false,
|
177 |
"special": true
|
178 |
},
|
179 |
"22": {
|
180 |
"content": "<unk_20>",
|
181 |
-
"lstrip":
|
182 |
"normalized": false,
|
183 |
-
"rstrip":
|
184 |
"single_word": false,
|
185 |
"special": true
|
186 |
},
|
187 |
"23": {
|
188 |
"content": "<unk_21>",
|
189 |
-
"lstrip":
|
190 |
"normalized": false,
|
191 |
-
"rstrip":
|
192 |
"single_word": false,
|
193 |
"special": true
|
194 |
},
|
195 |
"24": {
|
196 |
"content": "<unk_22>",
|
197 |
-
"lstrip":
|
198 |
"normalized": false,
|
199 |
-
"rstrip":
|
200 |
"single_word": false,
|
201 |
"special": true
|
202 |
},
|
203 |
"25": {
|
204 |
"content": "<unk_23>",
|
205 |
-
"lstrip":
|
206 |
"normalized": false,
|
207 |
-
"rstrip":
|
208 |
"single_word": false,
|
209 |
"special": true
|
210 |
},
|
211 |
"26": {
|
212 |
"content": "<unk_24>",
|
213 |
-
"lstrip":
|
214 |
"normalized": false,
|
215 |
-
"rstrip":
|
216 |
"single_word": false,
|
217 |
"special": true
|
218 |
},
|
219 |
"27": {
|
220 |
"content": "<unk_25>",
|
221 |
-
"lstrip":
|
222 |
"normalized": false,
|
223 |
-
"rstrip":
|
224 |
"single_word": false,
|
225 |
"special": true
|
226 |
},
|
227 |
"28": {
|
228 |
"content": "<unk_26>",
|
229 |
-
"lstrip":
|
230 |
"normalized": false,
|
231 |
-
"rstrip":
|
232 |
"single_word": false,
|
233 |
"special": true
|
234 |
},
|
235 |
"29": {
|
236 |
"content": "<unk_27>",
|
237 |
-
"lstrip":
|
238 |
"normalized": false,
|
239 |
-
"rstrip":
|
240 |
"single_word": false,
|
241 |
"special": true
|
242 |
},
|
243 |
"30": {
|
244 |
"content": "<unk_28>",
|
245 |
-
"lstrip":
|
246 |
"normalized": false,
|
247 |
-
"rstrip":
|
248 |
"single_word": false,
|
249 |
"special": true
|
250 |
},
|
251 |
"31": {
|
252 |
"content": "<unk_29>",
|
253 |
-
"lstrip":
|
254 |
"normalized": false,
|
255 |
-
"rstrip":
|
256 |
"single_word": false,
|
257 |
"special": true
|
258 |
},
|
259 |
"32": {
|
260 |
"content": "<unk_30>",
|
261 |
-
"lstrip":
|
262 |
"normalized": false,
|
263 |
-
"rstrip":
|
264 |
"single_word": false,
|
265 |
"special": true
|
266 |
},
|
267 |
"33": {
|
268 |
"content": "<unk_31>",
|
269 |
-
"lstrip":
|
270 |
"normalized": false,
|
271 |
-
"rstrip":
|
272 |
"single_word": false,
|
273 |
"special": true
|
274 |
},
|
275 |
"34": {
|
276 |
"content": "<unk_32>",
|
277 |
-
"lstrip":
|
278 |
"normalized": false,
|
279 |
-
"rstrip":
|
280 |
"single_word": false,
|
281 |
"special": true
|
282 |
},
|
283 |
"35": {
|
284 |
"content": "<unk_33>",
|
285 |
-
"lstrip":
|
286 |
"normalized": false,
|
287 |
-
"rstrip":
|
288 |
"single_word": false,
|
289 |
"special": true
|
290 |
},
|
291 |
"36": {
|
292 |
"content": "<unk_34>",
|
293 |
-
"lstrip":
|
294 |
"normalized": false,
|
295 |
-
"rstrip":
|
296 |
"single_word": false,
|
297 |
"special": true
|
298 |
},
|
299 |
"37": {
|
300 |
"content": "<unk_35>",
|
301 |
-
"lstrip":
|
302 |
"normalized": false,
|
303 |
-
"rstrip":
|
304 |
"single_word": false,
|
305 |
"special": true
|
306 |
},
|
307 |
"38": {
|
308 |
"content": "<unk_36>",
|
309 |
-
"lstrip":
|
310 |
"normalized": false,
|
311 |
-
"rstrip":
|
312 |
"single_word": false,
|
313 |
"special": true
|
314 |
},
|
315 |
"39": {
|
316 |
"content": "<unk_37>",
|
317 |
-
"lstrip":
|
318 |
"normalized": false,
|
319 |
-
"rstrip":
|
320 |
"single_word": false,
|
321 |
"special": true
|
322 |
},
|
323 |
"40": {
|
324 |
"content": "<unk_38>",
|
325 |
-
"lstrip":
|
326 |
"normalized": false,
|
327 |
-
"rstrip":
|
328 |
"single_word": false,
|
329 |
"special": true
|
330 |
},
|
331 |
"41": {
|
332 |
"content": "<unk_39>",
|
333 |
-
"lstrip":
|
334 |
"normalized": false,
|
335 |
-
"rstrip":
|
336 |
"single_word": false,
|
337 |
"special": true
|
338 |
},
|
339 |
"42": {
|
340 |
"content": "<unk_40>",
|
341 |
-
"lstrip":
|
342 |
"normalized": false,
|
343 |
-
"rstrip":
|
344 |
"single_word": false,
|
345 |
"special": true
|
346 |
},
|
347 |
"43": {
|
348 |
"content": "<unk_41>",
|
349 |
-
"lstrip":
|
350 |
"normalized": false,
|
351 |
-
"rstrip":
|
352 |
"single_word": false,
|
353 |
"special": true
|
354 |
},
|
355 |
"44": {
|
356 |
"content": "<unk_42>",
|
357 |
-
"lstrip":
|
358 |
"normalized": false,
|
359 |
-
"rstrip":
|
360 |
"single_word": false,
|
361 |
"special": true
|
362 |
},
|
363 |
"45": {
|
364 |
"content": "<unk_43>",
|
365 |
-
"lstrip":
|
366 |
"normalized": false,
|
367 |
-
"rstrip":
|
368 |
"single_word": false,
|
369 |
"special": true
|
370 |
},
|
371 |
"46": {
|
372 |
"content": "<unk_44>",
|
373 |
-
"lstrip":
|
374 |
"normalized": false,
|
375 |
-
"rstrip":
|
376 |
"single_word": false,
|
377 |
"special": true
|
378 |
},
|
379 |
"47": {
|
380 |
"content": "<unk_45>",
|
381 |
-
"lstrip":
|
382 |
"normalized": false,
|
383 |
-
"rstrip":
|
384 |
"single_word": false,
|
385 |
"special": true
|
386 |
},
|
387 |
"48": {
|
388 |
"content": "<unk_46>",
|
389 |
-
"lstrip":
|
390 |
"normalized": false,
|
391 |
-
"rstrip":
|
392 |
"single_word": false,
|
393 |
"special": true
|
394 |
},
|
395 |
"49": {
|
396 |
"content": "<unk_47>",
|
397 |
-
"lstrip":
|
398 |
"normalized": false,
|
399 |
-
"rstrip":
|
400 |
"single_word": false,
|
401 |
"special": true
|
402 |
},
|
403 |
"50": {
|
404 |
"content": "<unk_48>",
|
405 |
-
"lstrip":
|
406 |
"normalized": false,
|
407 |
-
"rstrip":
|
408 |
"single_word": false,
|
409 |
"special": true
|
410 |
},
|
411 |
"51": {
|
412 |
"content": "<unk_49>",
|
413 |
-
"lstrip":
|
414 |
"normalized": false,
|
415 |
-
"rstrip":
|
416 |
"single_word": false,
|
417 |
"special": true
|
418 |
},
|
419 |
"52": {
|
420 |
"content": "<unk_50>",
|
421 |
-
"lstrip":
|
422 |
"normalized": false,
|
423 |
-
"rstrip":
|
424 |
"single_word": false,
|
425 |
"special": true
|
426 |
},
|
427 |
"53": {
|
428 |
"content": "<unk_51>",
|
429 |
-
"lstrip":
|
430 |
"normalized": false,
|
431 |
-
"rstrip":
|
432 |
"single_word": false,
|
433 |
"special": true
|
434 |
},
|
435 |
"54": {
|
436 |
"content": "<unk_52>",
|
437 |
-
"lstrip":
|
438 |
"normalized": false,
|
439 |
-
"rstrip":
|
440 |
"single_word": false,
|
441 |
"special": true
|
442 |
},
|
443 |
"55": {
|
444 |
"content": "<unk_53>",
|
445 |
-
"lstrip":
|
446 |
"normalized": false,
|
447 |
-
"rstrip":
|
448 |
"single_word": false,
|
449 |
"special": true
|
450 |
},
|
451 |
"56": {
|
452 |
"content": "<unk_54>",
|
453 |
-
"lstrip":
|
454 |
"normalized": false,
|
455 |
-
"rstrip":
|
456 |
"single_word": false,
|
457 |
"special": true
|
458 |
},
|
459 |
"57": {
|
460 |
"content": "<unk_55>",
|
461 |
-
"lstrip":
|
462 |
"normalized": false,
|
463 |
-
"rstrip":
|
464 |
"single_word": false,
|
465 |
"special": true
|
466 |
},
|
467 |
"58": {
|
468 |
"content": "<unk_56>",
|
469 |
-
"lstrip":
|
470 |
"normalized": false,
|
471 |
-
"rstrip":
|
472 |
"single_word": false,
|
473 |
"special": true
|
474 |
},
|
475 |
"59": {
|
476 |
"content": "<unk_57>",
|
477 |
-
"lstrip":
|
478 |
"normalized": false,
|
479 |
-
"rstrip":
|
480 |
"single_word": false,
|
481 |
"special": true
|
482 |
},
|
483 |
"60": {
|
484 |
"content": "<unk_58>",
|
485 |
-
"lstrip":
|
486 |
"normalized": false,
|
487 |
-
"rstrip":
|
488 |
"single_word": false,
|
489 |
"special": true
|
490 |
},
|
491 |
"61": {
|
492 |
"content": "<unk_59>",
|
493 |
-
"lstrip":
|
494 |
"normalized": false,
|
495 |
-
"rstrip":
|
496 |
"single_word": false,
|
497 |
"special": true
|
498 |
},
|
499 |
"62": {
|
500 |
"content": "<unk_60>",
|
501 |
-
"lstrip":
|
502 |
"normalized": false,
|
503 |
-
"rstrip":
|
504 |
"single_word": false,
|
505 |
"special": true
|
506 |
},
|
507 |
"63": {
|
508 |
"content": "<unk_61>",
|
509 |
-
"lstrip":
|
510 |
"normalized": false,
|
511 |
-
"rstrip":
|
512 |
"single_word": false,
|
513 |
"special": true
|
514 |
},
|
515 |
"64": {
|
516 |
"content": "<unk_62>",
|
517 |
-
"lstrip":
|
518 |
"normalized": false,
|
519 |
-
"rstrip":
|
520 |
"single_word": false,
|
521 |
"special": true
|
522 |
},
|
523 |
"65": {
|
524 |
"content": "<unk_63>",
|
525 |
-
"lstrip":
|
526 |
"normalized": false,
|
527 |
-
"rstrip":
|
528 |
"single_word": false,
|
529 |
"special": true
|
530 |
},
|
531 |
"66": {
|
532 |
"content": "<unk_64>",
|
533 |
-
"lstrip":
|
534 |
"normalized": false,
|
535 |
-
"rstrip":
|
536 |
"single_word": false,
|
537 |
"special": true
|
538 |
},
|
539 |
"67": {
|
540 |
"content": "<unk_65>",
|
541 |
-
"lstrip":
|
542 |
"normalized": false,
|
543 |
-
"rstrip":
|
544 |
"single_word": false,
|
545 |
"special": true
|
546 |
},
|
547 |
"68": {
|
548 |
"content": "<unk_66>",
|
549 |
-
"lstrip":
|
550 |
"normalized": false,
|
551 |
-
"rstrip":
|
552 |
"single_word": false,
|
553 |
"special": true
|
554 |
},
|
555 |
"69": {
|
556 |
"content": "<unk_67>",
|
557 |
-
"lstrip":
|
558 |
"normalized": false,
|
559 |
-
"rstrip":
|
560 |
"single_word": false,
|
561 |
"special": true
|
562 |
},
|
563 |
"70": {
|
564 |
"content": "<unk_68>",
|
565 |
-
"lstrip":
|
566 |
"normalized": false,
|
567 |
-
"rstrip":
|
568 |
"single_word": false,
|
569 |
"special": true
|
570 |
},
|
571 |
"71": {
|
572 |
"content": "<unk_69>",
|
573 |
-
"lstrip":
|
574 |
"normalized": false,
|
575 |
-
"rstrip":
|
576 |
"single_word": false,
|
577 |
"special": true
|
578 |
},
|
579 |
"72": {
|
580 |
"content": "<unk_70>",
|
581 |
-
"lstrip":
|
582 |
"normalized": false,
|
583 |
-
"rstrip":
|
584 |
"single_word": false,
|
585 |
"special": true
|
586 |
},
|
587 |
"73": {
|
588 |
"content": "<unk_71>",
|
589 |
-
"lstrip":
|
590 |
"normalized": false,
|
591 |
-
"rstrip":
|
592 |
"single_word": false,
|
593 |
"special": true
|
594 |
},
|
595 |
"74": {
|
596 |
"content": "<unk_72>",
|
597 |
-
"lstrip":
|
598 |
"normalized": false,
|
599 |
-
"rstrip":
|
600 |
"single_word": false,
|
601 |
"special": true
|
602 |
},
|
603 |
"75": {
|
604 |
"content": "<unk_73>",
|
605 |
-
"lstrip":
|
606 |
"normalized": false,
|
607 |
-
"rstrip":
|
608 |
"single_word": false,
|
609 |
"special": true
|
610 |
},
|
611 |
"76": {
|
612 |
"content": "<unk_74>",
|
613 |
-
"lstrip":
|
614 |
"normalized": false,
|
615 |
-
"rstrip":
|
616 |
"single_word": false,
|
617 |
"special": true
|
618 |
},
|
619 |
"77": {
|
620 |
"content": "<unk_75>",
|
621 |
-
"lstrip":
|
622 |
"normalized": false,
|
623 |
-
"rstrip":
|
624 |
"single_word": false,
|
625 |
"special": true
|
626 |
},
|
627 |
"78": {
|
628 |
"content": "<unk_76>",
|
629 |
-
"lstrip":
|
630 |
"normalized": false,
|
631 |
-
"rstrip":
|
632 |
"single_word": false,
|
633 |
"special": true
|
634 |
},
|
635 |
"79": {
|
636 |
"content": "<unk_77>",
|
637 |
-
"lstrip":
|
638 |
"normalized": false,
|
639 |
-
"rstrip":
|
640 |
"single_word": false,
|
641 |
"special": true
|
642 |
},
|
643 |
"80": {
|
644 |
"content": "<unk_78>",
|
645 |
-
"lstrip":
|
646 |
"normalized": false,
|
647 |
-
"rstrip":
|
648 |
"single_word": false,
|
649 |
"special": true
|
650 |
},
|
651 |
"81": {
|
652 |
"content": "<unk_79>",
|
653 |
-
"lstrip":
|
654 |
"normalized": false,
|
655 |
-
"rstrip":
|
656 |
"single_word": false,
|
657 |
"special": true
|
658 |
},
|
659 |
"82": {
|
660 |
"content": "<unk_80>",
|
661 |
-
"lstrip":
|
662 |
"normalized": false,
|
663 |
-
"rstrip":
|
664 |
"single_word": false,
|
665 |
"special": true
|
666 |
},
|
667 |
"83": {
|
668 |
"content": "<unk_81>",
|
669 |
-
"lstrip":
|
670 |
"normalized": false,
|
671 |
-
"rstrip":
|
672 |
"single_word": false,
|
673 |
"special": true
|
674 |
},
|
675 |
"84": {
|
676 |
"content": "<unk_82>",
|
677 |
-
"lstrip":
|
678 |
"normalized": false,
|
679 |
-
"rstrip":
|
680 |
"single_word": false,
|
681 |
"special": true
|
682 |
},
|
683 |
"85": {
|
684 |
"content": "<unk_83>",
|
685 |
-
"lstrip":
|
686 |
"normalized": false,
|
687 |
-
"rstrip":
|
688 |
"single_word": false,
|
689 |
"special": true
|
690 |
},
|
691 |
"86": {
|
692 |
"content": "<unk_84>",
|
693 |
-
"lstrip":
|
694 |
"normalized": false,
|
695 |
-
"rstrip":
|
696 |
"single_word": false,
|
697 |
"special": true
|
698 |
},
|
699 |
"87": {
|
700 |
"content": "<unk_85>",
|
701 |
-
"lstrip":
|
702 |
"normalized": false,
|
703 |
-
"rstrip":
|
704 |
"single_word": false,
|
705 |
"special": true
|
706 |
},
|
707 |
"88": {
|
708 |
"content": "<unk_86>",
|
709 |
-
"lstrip":
|
710 |
"normalized": false,
|
711 |
-
"rstrip":
|
712 |
"single_word": false,
|
713 |
"special": true
|
714 |
},
|
715 |
"89": {
|
716 |
"content": "<unk_87>",
|
717 |
-
"lstrip":
|
718 |
"normalized": false,
|
719 |
-
"rstrip":
|
720 |
"single_word": false,
|
721 |
"special": true
|
722 |
},
|
723 |
"90": {
|
724 |
"content": "<unk_88>",
|
725 |
-
"lstrip":
|
726 |
"normalized": false,
|
727 |
-
"rstrip":
|
728 |
"single_word": false,
|
729 |
"special": true
|
730 |
},
|
731 |
"91": {
|
732 |
"content": "<unk_89>",
|
733 |
-
"lstrip":
|
734 |
"normalized": false,
|
735 |
-
"rstrip":
|
736 |
"single_word": false,
|
737 |
"special": true
|
738 |
},
|
739 |
"92": {
|
740 |
"content": "<unk_90>",
|
741 |
-
"lstrip":
|
742 |
"normalized": false,
|
743 |
-
"rstrip":
|
744 |
"single_word": false,
|
745 |
"special": true
|
746 |
},
|
747 |
"93": {
|
748 |
"content": "<unk_91>",
|
749 |
-
"lstrip":
|
750 |
"normalized": false,
|
751 |
-
"rstrip":
|
752 |
"single_word": false,
|
753 |
"special": true
|
754 |
},
|
755 |
"94": {
|
756 |
"content": "<unk_92>",
|
757 |
-
"lstrip":
|
758 |
"normalized": false,
|
759 |
-
"rstrip":
|
760 |
"single_word": false,
|
761 |
"special": true
|
762 |
},
|
763 |
"95": {
|
764 |
"content": "<unk_93>",
|
765 |
-
"lstrip":
|
766 |
"normalized": false,
|
767 |
-
"rstrip":
|
768 |
"single_word": false,
|
769 |
"special": true
|
770 |
},
|
771 |
"96": {
|
772 |
"content": "<unk_94>",
|
773 |
-
"lstrip":
|
774 |
"normalized": false,
|
775 |
-
"rstrip":
|
776 |
"single_word": false,
|
777 |
"special": true
|
778 |
},
|
779 |
"97": {
|
780 |
"content": "<unk_95>",
|
781 |
-
"lstrip":
|
782 |
"normalized": false,
|
783 |
-
"rstrip":
|
784 |
"single_word": false,
|
785 |
"special": true
|
786 |
},
|
787 |
"98": {
|
788 |
"content": "<unk_96>",
|
789 |
-
"lstrip":
|
790 |
"normalized": false,
|
791 |
-
"rstrip":
|
792 |
"single_word": false,
|
793 |
"special": true
|
794 |
},
|
795 |
"99": {
|
796 |
"content": "<unk_97>",
|
797 |
-
"lstrip":
|
798 |
"normalized": false,
|
799 |
-
"rstrip":
|
800 |
"single_word": false,
|
801 |
"special": true
|
802 |
},
|
803 |
"100": {
|
804 |
"content": "<unk_98>",
|
805 |
-
"lstrip":
|
806 |
"normalized": false,
|
807 |
-
"rstrip":
|
808 |
"single_word": false,
|
809 |
"special": true
|
810 |
},
|
811 |
"101": {
|
812 |
"content": "<unk_99>",
|
813 |
-
"lstrip":
|
814 |
"normalized": false,
|
815 |
-
"rstrip":
|
816 |
"single_word": false,
|
817 |
"special": true
|
818 |
},
|
819 |
"102": {
|
820 |
"content": "<unk_100>",
|
821 |
-
"lstrip":
|
822 |
"normalized": false,
|
823 |
-
"rstrip":
|
824 |
"single_word": false,
|
825 |
"special": true
|
826 |
},
|
827 |
"103": {
|
828 |
"content": "<unk_101>",
|
829 |
-
"lstrip":
|
830 |
"normalized": false,
|
831 |
-
"rstrip":
|
832 |
"single_word": false,
|
833 |
"special": true
|
834 |
},
|
835 |
"104": {
|
836 |
"content": "<unk_102>",
|
837 |
-
"lstrip":
|
838 |
"normalized": false,
|
839 |
-
"rstrip":
|
840 |
"single_word": false,
|
841 |
"special": true
|
842 |
},
|
|
|
18 |
},
|
19 |
"2": {
|
20 |
"content": "<mask_1>",
|
21 |
+
"lstrip": false,
|
22 |
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
"single_word": false,
|
25 |
"special": true
|
26 |
},
|
|
|
34 |
},
|
35 |
"4": {
|
36 |
"content": "<unk_2>",
|
37 |
+
"lstrip": false,
|
38 |
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
"single_word": false,
|
41 |
"special": true
|
42 |
},
|
43 |
"5": {
|
44 |
"content": "<unk_3>",
|
45 |
+
"lstrip": false,
|
46 |
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
"single_word": false,
|
49 |
"special": true
|
50 |
},
|
51 |
"6": {
|
52 |
"content": "<unk_4>",
|
53 |
+
"lstrip": false,
|
54 |
"normalized": false,
|
55 |
+
"rstrip": false,
|
56 |
"single_word": false,
|
57 |
"special": true
|
58 |
},
|
59 |
"7": {
|
60 |
"content": "<unk_5>",
|
61 |
+
"lstrip": false,
|
62 |
"normalized": false,
|
63 |
+
"rstrip": false,
|
64 |
"single_word": false,
|
65 |
"special": true
|
66 |
},
|
67 |
"8": {
|
68 |
"content": "<unk_6>",
|
69 |
+
"lstrip": false,
|
70 |
"normalized": false,
|
71 |
+
"rstrip": false,
|
72 |
"single_word": false,
|
73 |
"special": true
|
74 |
},
|
75 |
"9": {
|
76 |
"content": "<unk_7>",
|
77 |
+
"lstrip": false,
|
78 |
"normalized": false,
|
79 |
+
"rstrip": false,
|
80 |
"single_word": false,
|
81 |
"special": true
|
82 |
},
|
83 |
"10": {
|
84 |
"content": "<unk_8>",
|
85 |
+
"lstrip": false,
|
86 |
"normalized": false,
|
87 |
+
"rstrip": false,
|
88 |
"single_word": false,
|
89 |
"special": true
|
90 |
},
|
91 |
"11": {
|
92 |
"content": "<unk_9>",
|
93 |
+
"lstrip": false,
|
94 |
"normalized": false,
|
95 |
+
"rstrip": false,
|
96 |
"single_word": false,
|
97 |
"special": true
|
98 |
},
|
99 |
"12": {
|
100 |
"content": "<unk_10>",
|
101 |
+
"lstrip": false,
|
102 |
"normalized": false,
|
103 |
+
"rstrip": false,
|
104 |
"single_word": false,
|
105 |
"special": true
|
106 |
},
|
107 |
"13": {
|
108 |
"content": "<unk_11>",
|
109 |
+
"lstrip": false,
|
110 |
"normalized": false,
|
111 |
+
"rstrip": false,
|
112 |
"single_word": false,
|
113 |
"special": true
|
114 |
},
|
115 |
"14": {
|
116 |
"content": "<unk_12>",
|
117 |
+
"lstrip": false,
|
118 |
"normalized": false,
|
119 |
+
"rstrip": false,
|
120 |
"single_word": false,
|
121 |
"special": true
|
122 |
},
|
123 |
"15": {
|
124 |
"content": "<unk_13>",
|
125 |
+
"lstrip": false,
|
126 |
"normalized": false,
|
127 |
+
"rstrip": false,
|
128 |
"single_word": false,
|
129 |
"special": true
|
130 |
},
|
131 |
"16": {
|
132 |
"content": "<unk_14>",
|
133 |
+
"lstrip": false,
|
134 |
"normalized": false,
|
135 |
+
"rstrip": false,
|
136 |
"single_word": false,
|
137 |
"special": true
|
138 |
},
|
139 |
"17": {
|
140 |
"content": "<unk_15>",
|
141 |
+
"lstrip": false,
|
142 |
"normalized": false,
|
143 |
+
"rstrip": false,
|
144 |
"single_word": false,
|
145 |
"special": true
|
146 |
},
|
147 |
"18": {
|
148 |
"content": "<unk_16>",
|
149 |
+
"lstrip": false,
|
150 |
"normalized": false,
|
151 |
+
"rstrip": false,
|
152 |
"single_word": false,
|
153 |
"special": true
|
154 |
},
|
155 |
"19": {
|
156 |
"content": "<unk_17>",
|
157 |
+
"lstrip": false,
|
158 |
"normalized": false,
|
159 |
+
"rstrip": false,
|
160 |
"single_word": false,
|
161 |
"special": true
|
162 |
},
|
163 |
"20": {
|
164 |
"content": "<unk_18>",
|
165 |
+
"lstrip": false,
|
166 |
"normalized": false,
|
167 |
+
"rstrip": false,
|
168 |
"single_word": false,
|
169 |
"special": true
|
170 |
},
|
171 |
"21": {
|
172 |
"content": "<unk_19>",
|
173 |
+
"lstrip": false,
|
174 |
"normalized": false,
|
175 |
+
"rstrip": false,
|
176 |
"single_word": false,
|
177 |
"special": true
|
178 |
},
|
179 |
"22": {
|
180 |
"content": "<unk_20>",
|
181 |
+
"lstrip": false,
|
182 |
"normalized": false,
|
183 |
+
"rstrip": false,
|
184 |
"single_word": false,
|
185 |
"special": true
|
186 |
},
|
187 |
"23": {
|
188 |
"content": "<unk_21>",
|
189 |
+
"lstrip": false,
|
190 |
"normalized": false,
|
191 |
+
"rstrip": false,
|
192 |
"single_word": false,
|
193 |
"special": true
|
194 |
},
|
195 |
"24": {
|
196 |
"content": "<unk_22>",
|
197 |
+
"lstrip": false,
|
198 |
"normalized": false,
|
199 |
+
"rstrip": false,
|
200 |
"single_word": false,
|
201 |
"special": true
|
202 |
},
|
203 |
"25": {
|
204 |
"content": "<unk_23>",
|
205 |
+
"lstrip": false,
|
206 |
"normalized": false,
|
207 |
+
"rstrip": false,
|
208 |
"single_word": false,
|
209 |
"special": true
|
210 |
},
|
211 |
"26": {
|
212 |
"content": "<unk_24>",
|
213 |
+
"lstrip": false,
|
214 |
"normalized": false,
|
215 |
+
"rstrip": false,
|
216 |
"single_word": false,
|
217 |
"special": true
|
218 |
},
|
219 |
"27": {
|
220 |
"content": "<unk_25>",
|
221 |
+
"lstrip": false,
|
222 |
"normalized": false,
|
223 |
+
"rstrip": false,
|
224 |
"single_word": false,
|
225 |
"special": true
|
226 |
},
|
227 |
"28": {
|
228 |
"content": "<unk_26>",
|
229 |
+
"lstrip": false,
|
230 |
"normalized": false,
|
231 |
+
"rstrip": false,
|
232 |
"single_word": false,
|
233 |
"special": true
|
234 |
},
|
235 |
"29": {
|
236 |
"content": "<unk_27>",
|
237 |
+
"lstrip": false,
|
238 |
"normalized": false,
|
239 |
+
"rstrip": false,
|
240 |
"single_word": false,
|
241 |
"special": true
|
242 |
},
|
243 |
"30": {
|
244 |
"content": "<unk_28>",
|
245 |
+
"lstrip": false,
|
246 |
"normalized": false,
|
247 |
+
"rstrip": false,
|
248 |
"single_word": false,
|
249 |
"special": true
|
250 |
},
|
251 |
"31": {
|
252 |
"content": "<unk_29>",
|
253 |
+
"lstrip": false,
|
254 |
"normalized": false,
|
255 |
+
"rstrip": false,
|
256 |
"single_word": false,
|
257 |
"special": true
|
258 |
},
|
259 |
"32": {
|
260 |
"content": "<unk_30>",
|
261 |
+
"lstrip": false,
|
262 |
"normalized": false,
|
263 |
+
"rstrip": false,
|
264 |
"single_word": false,
|
265 |
"special": true
|
266 |
},
|
267 |
"33": {
|
268 |
"content": "<unk_31>",
|
269 |
+
"lstrip": false,
|
270 |
"normalized": false,
|
271 |
+
"rstrip": false,
|
272 |
"single_word": false,
|
273 |
"special": true
|
274 |
},
|
275 |
"34": {
|
276 |
"content": "<unk_32>",
|
277 |
+
"lstrip": false,
|
278 |
"normalized": false,
|
279 |
+
"rstrip": false,
|
280 |
"single_word": false,
|
281 |
"special": true
|
282 |
},
|
283 |
"35": {
|
284 |
"content": "<unk_33>",
|
285 |
+
"lstrip": false,
|
286 |
"normalized": false,
|
287 |
+
"rstrip": false,
|
288 |
"single_word": false,
|
289 |
"special": true
|
290 |
},
|
291 |
"36": {
|
292 |
"content": "<unk_34>",
|
293 |
+
"lstrip": false,
|
294 |
"normalized": false,
|
295 |
+
"rstrip": false,
|
296 |
"single_word": false,
|
297 |
"special": true
|
298 |
},
|
299 |
"37": {
|
300 |
"content": "<unk_35>",
|
301 |
+
"lstrip": false,
|
302 |
"normalized": false,
|
303 |
+
"rstrip": false,
|
304 |
"single_word": false,
|
305 |
"special": true
|
306 |
},
|
307 |
"38": {
|
308 |
"content": "<unk_36>",
|
309 |
+
"lstrip": false,
|
310 |
"normalized": false,
|
311 |
+
"rstrip": false,
|
312 |
"single_word": false,
|
313 |
"special": true
|
314 |
},
|
315 |
"39": {
|
316 |
"content": "<unk_37>",
|
317 |
+
"lstrip": false,
|
318 |
"normalized": false,
|
319 |
+
"rstrip": false,
|
320 |
"single_word": false,
|
321 |
"special": true
|
322 |
},
|
323 |
"40": {
|
324 |
"content": "<unk_38>",
|
325 |
+
"lstrip": false,
|
326 |
"normalized": false,
|
327 |
+
"rstrip": false,
|
328 |
"single_word": false,
|
329 |
"special": true
|
330 |
},
|
331 |
"41": {
|
332 |
"content": "<unk_39>",
|
333 |
+
"lstrip": false,
|
334 |
"normalized": false,
|
335 |
+
"rstrip": false,
|
336 |
"single_word": false,
|
337 |
"special": true
|
338 |
},
|
339 |
"42": {
|
340 |
"content": "<unk_40>",
|
341 |
+
"lstrip": false,
|
342 |
"normalized": false,
|
343 |
+
"rstrip": false,
|
344 |
"single_word": false,
|
345 |
"special": true
|
346 |
},
|
347 |
"43": {
|
348 |
"content": "<unk_41>",
|
349 |
+
"lstrip": false,
|
350 |
"normalized": false,
|
351 |
+
"rstrip": false,
|
352 |
"single_word": false,
|
353 |
"special": true
|
354 |
},
|
355 |
"44": {
|
356 |
"content": "<unk_42>",
|
357 |
+
"lstrip": false,
|
358 |
"normalized": false,
|
359 |
+
"rstrip": false,
|
360 |
"single_word": false,
|
361 |
"special": true
|
362 |
},
|
363 |
"45": {
|
364 |
"content": "<unk_43>",
|
365 |
+
"lstrip": false,
|
366 |
"normalized": false,
|
367 |
+
"rstrip": false,
|
368 |
"single_word": false,
|
369 |
"special": true
|
370 |
},
|
371 |
"46": {
|
372 |
"content": "<unk_44>",
|
373 |
+
"lstrip": false,
|
374 |
"normalized": false,
|
375 |
+
"rstrip": false,
|
376 |
"single_word": false,
|
377 |
"special": true
|
378 |
},
|
379 |
"47": {
|
380 |
"content": "<unk_45>",
|
381 |
+
"lstrip": false,
|
382 |
"normalized": false,
|
383 |
+
"rstrip": false,
|
384 |
"single_word": false,
|
385 |
"special": true
|
386 |
},
|
387 |
"48": {
|
388 |
"content": "<unk_46>",
|
389 |
+
"lstrip": false,
|
390 |
"normalized": false,
|
391 |
+
"rstrip": false,
|
392 |
"single_word": false,
|
393 |
"special": true
|
394 |
},
|
395 |
"49": {
|
396 |
"content": "<unk_47>",
|
397 |
+
"lstrip": false,
|
398 |
"normalized": false,
|
399 |
+
"rstrip": false,
|
400 |
"single_word": false,
|
401 |
"special": true
|
402 |
},
|
403 |
"50": {
|
404 |
"content": "<unk_48>",
|
405 |
+
"lstrip": false,
|
406 |
"normalized": false,
|
407 |
+
"rstrip": false,
|
408 |
"single_word": false,
|
409 |
"special": true
|
410 |
},
|
411 |
"51": {
|
412 |
"content": "<unk_49>",
|
413 |
+
"lstrip": false,
|
414 |
"normalized": false,
|
415 |
+
"rstrip": false,
|
416 |
"single_word": false,
|
417 |
"special": true
|
418 |
},
|
419 |
"52": {
|
420 |
"content": "<unk_50>",
|
421 |
+
"lstrip": false,
|
422 |
"normalized": false,
|
423 |
+
"rstrip": false,
|
424 |
"single_word": false,
|
425 |
"special": true
|
426 |
},
|
427 |
"53": {
|
428 |
"content": "<unk_51>",
|
429 |
+
"lstrip": false,
|
430 |
"normalized": false,
|
431 |
+
"rstrip": false,
|
432 |
"single_word": false,
|
433 |
"special": true
|
434 |
},
|
435 |
"54": {
|
436 |
"content": "<unk_52>",
|
437 |
+
"lstrip": false,
|
438 |
"normalized": false,
|
439 |
+
"rstrip": false,
|
440 |
"single_word": false,
|
441 |
"special": true
|
442 |
},
|
443 |
"55": {
|
444 |
"content": "<unk_53>",
|
445 |
+
"lstrip": false,
|
446 |
"normalized": false,
|
447 |
+
"rstrip": false,
|
448 |
"single_word": false,
|
449 |
"special": true
|
450 |
},
|
451 |
"56": {
|
452 |
"content": "<unk_54>",
|
453 |
+
"lstrip": false,
|
454 |
"normalized": false,
|
455 |
+
"rstrip": false,
|
456 |
"single_word": false,
|
457 |
"special": true
|
458 |
},
|
459 |
"57": {
|
460 |
"content": "<unk_55>",
|
461 |
+
"lstrip": false,
|
462 |
"normalized": false,
|
463 |
+
"rstrip": false,
|
464 |
"single_word": false,
|
465 |
"special": true
|
466 |
},
|
467 |
"58": {
|
468 |
"content": "<unk_56>",
|
469 |
+
"lstrip": false,
|
470 |
"normalized": false,
|
471 |
+
"rstrip": false,
|
472 |
"single_word": false,
|
473 |
"special": true
|
474 |
},
|
475 |
"59": {
|
476 |
"content": "<unk_57>",
|
477 |
+
"lstrip": false,
|
478 |
"normalized": false,
|
479 |
+
"rstrip": false,
|
480 |
"single_word": false,
|
481 |
"special": true
|
482 |
},
|
483 |
"60": {
|
484 |
"content": "<unk_58>",
|
485 |
+
"lstrip": false,
|
486 |
"normalized": false,
|
487 |
+
"rstrip": false,
|
488 |
"single_word": false,
|
489 |
"special": true
|
490 |
},
|
491 |
"61": {
|
492 |
"content": "<unk_59>",
|
493 |
+
"lstrip": false,
|
494 |
"normalized": false,
|
495 |
+
"rstrip": false,
|
496 |
"single_word": false,
|
497 |
"special": true
|
498 |
},
|
499 |
"62": {
|
500 |
"content": "<unk_60>",
|
501 |
+
"lstrip": false,
|
502 |
"normalized": false,
|
503 |
+
"rstrip": false,
|
504 |
"single_word": false,
|
505 |
"special": true
|
506 |
},
|
507 |
"63": {
|
508 |
"content": "<unk_61>",
|
509 |
+
"lstrip": false,
|
510 |
"normalized": false,
|
511 |
+
"rstrip": false,
|
512 |
"single_word": false,
|
513 |
"special": true
|
514 |
},
|
515 |
"64": {
|
516 |
"content": "<unk_62>",
|
517 |
+
"lstrip": false,
|
518 |
"normalized": false,
|
519 |
+
"rstrip": false,
|
520 |
"single_word": false,
|
521 |
"special": true
|
522 |
},
|
523 |
"65": {
|
524 |
"content": "<unk_63>",
|
525 |
+
"lstrip": false,
|
526 |
"normalized": false,
|
527 |
+
"rstrip": false,
|
528 |
"single_word": false,
|
529 |
"special": true
|
530 |
},
|
531 |
"66": {
|
532 |
"content": "<unk_64>",
|
533 |
+
"lstrip": false,
|
534 |
"normalized": false,
|
535 |
+
"rstrip": false,
|
536 |
"single_word": false,
|
537 |
"special": true
|
538 |
},
|
539 |
"67": {
|
540 |
"content": "<unk_65>",
|
541 |
+
"lstrip": false,
|
542 |
"normalized": false,
|
543 |
+
"rstrip": false,
|
544 |
"single_word": false,
|
545 |
"special": true
|
546 |
},
|
547 |
"68": {
|
548 |
"content": "<unk_66>",
|
549 |
+
"lstrip": false,
|
550 |
"normalized": false,
|
551 |
+
"rstrip": false,
|
552 |
"single_word": false,
|
553 |
"special": true
|
554 |
},
|
555 |
"69": {
|
556 |
"content": "<unk_67>",
|
557 |
+
"lstrip": false,
|
558 |
"normalized": false,
|
559 |
+
"rstrip": false,
|
560 |
"single_word": false,
|
561 |
"special": true
|
562 |
},
|
563 |
"70": {
|
564 |
"content": "<unk_68>",
|
565 |
+
"lstrip": false,
|
566 |
"normalized": false,
|
567 |
+
"rstrip": false,
|
568 |
"single_word": false,
|
569 |
"special": true
|
570 |
},
|
571 |
"71": {
|
572 |
"content": "<unk_69>",
|
573 |
+
"lstrip": false,
|
574 |
"normalized": false,
|
575 |
+
"rstrip": false,
|
576 |
"single_word": false,
|
577 |
"special": true
|
578 |
},
|
579 |
"72": {
|
580 |
"content": "<unk_70>",
|
581 |
+
"lstrip": false,
|
582 |
"normalized": false,
|
583 |
+
"rstrip": false,
|
584 |
"single_word": false,
|
585 |
"special": true
|
586 |
},
|
587 |
"73": {
|
588 |
"content": "<unk_71>",
|
589 |
+
"lstrip": false,
|
590 |
"normalized": false,
|
591 |
+
"rstrip": false,
|
592 |
"single_word": false,
|
593 |
"special": true
|
594 |
},
|
595 |
"74": {
|
596 |
"content": "<unk_72>",
|
597 |
+
"lstrip": false,
|
598 |
"normalized": false,
|
599 |
+
"rstrip": false,
|
600 |
"single_word": false,
|
601 |
"special": true
|
602 |
},
|
603 |
"75": {
|
604 |
"content": "<unk_73>",
|
605 |
+
"lstrip": false,
|
606 |
"normalized": false,
|
607 |
+
"rstrip": false,
|
608 |
"single_word": false,
|
609 |
"special": true
|
610 |
},
|
611 |
"76": {
|
612 |
"content": "<unk_74>",
|
613 |
+
"lstrip": false,
|
614 |
"normalized": false,
|
615 |
+
"rstrip": false,
|
616 |
"single_word": false,
|
617 |
"special": true
|
618 |
},
|
619 |
"77": {
|
620 |
"content": "<unk_75>",
|
621 |
+
"lstrip": false,
|
622 |
"normalized": false,
|
623 |
+
"rstrip": false,
|
624 |
"single_word": false,
|
625 |
"special": true
|
626 |
},
|
627 |
"78": {
|
628 |
"content": "<unk_76>",
|
629 |
+
"lstrip": false,
|
630 |
"normalized": false,
|
631 |
+
"rstrip": false,
|
632 |
"single_word": false,
|
633 |
"special": true
|
634 |
},
|
635 |
"79": {
|
636 |
"content": "<unk_77>",
|
637 |
+
"lstrip": false,
|
638 |
"normalized": false,
|
639 |
+
"rstrip": false,
|
640 |
"single_word": false,
|
641 |
"special": true
|
642 |
},
|
643 |
"80": {
|
644 |
"content": "<unk_78>",
|
645 |
+
"lstrip": false,
|
646 |
"normalized": false,
|
647 |
+
"rstrip": false,
|
648 |
"single_word": false,
|
649 |
"special": true
|
650 |
},
|
651 |
"81": {
|
652 |
"content": "<unk_79>",
|
653 |
+
"lstrip": false,
|
654 |
"normalized": false,
|
655 |
+
"rstrip": false,
|
656 |
"single_word": false,
|
657 |
"special": true
|
658 |
},
|
659 |
"82": {
|
660 |
"content": "<unk_80>",
|
661 |
+
"lstrip": false,
|
662 |
"normalized": false,
|
663 |
+
"rstrip": false,
|
664 |
"single_word": false,
|
665 |
"special": true
|
666 |
},
|
667 |
"83": {
|
668 |
"content": "<unk_81>",
|
669 |
+
"lstrip": false,
|
670 |
"normalized": false,
|
671 |
+
"rstrip": false,
|
672 |
"single_word": false,
|
673 |
"special": true
|
674 |
},
|
675 |
"84": {
|
676 |
"content": "<unk_82>",
|
677 |
+
"lstrip": false,
|
678 |
"normalized": false,
|
679 |
+
"rstrip": false,
|
680 |
"single_word": false,
|
681 |
"special": true
|
682 |
},
|
683 |
"85": {
|
684 |
"content": "<unk_83>",
|
685 |
+
"lstrip": false,
|
686 |
"normalized": false,
|
687 |
+
"rstrip": false,
|
688 |
"single_word": false,
|
689 |
"special": true
|
690 |
},
|
691 |
"86": {
|
692 |
"content": "<unk_84>",
|
693 |
+
"lstrip": false,
|
694 |
"normalized": false,
|
695 |
+
"rstrip": false,
|
696 |
"single_word": false,
|
697 |
"special": true
|
698 |
},
|
699 |
"87": {
|
700 |
"content": "<unk_85>",
|
701 |
+
"lstrip": false,
|
702 |
"normalized": false,
|
703 |
+
"rstrip": false,
|
704 |
"single_word": false,
|
705 |
"special": true
|
706 |
},
|
707 |
"88": {
|
708 |
"content": "<unk_86>",
|
709 |
+
"lstrip": false,
|
710 |
"normalized": false,
|
711 |
+
"rstrip": false,
|
712 |
"single_word": false,
|
713 |
"special": true
|
714 |
},
|
715 |
"89": {
|
716 |
"content": "<unk_87>",
|
717 |
+
"lstrip": false,
|
718 |
"normalized": false,
|
719 |
+
"rstrip": false,
|
720 |
"single_word": false,
|
721 |
"special": true
|
722 |
},
|
723 |
"90": {
|
724 |
"content": "<unk_88>",
|
725 |
+
"lstrip": false,
|
726 |
"normalized": false,
|
727 |
+
"rstrip": false,
|
728 |
"single_word": false,
|
729 |
"special": true
|
730 |
},
|
731 |
"91": {
|
732 |
"content": "<unk_89>",
|
733 |
+
"lstrip": false,
|
734 |
"normalized": false,
|
735 |
+
"rstrip": false,
|
736 |
"single_word": false,
|
737 |
"special": true
|
738 |
},
|
739 |
"92": {
|
740 |
"content": "<unk_90>",
|
741 |
+
"lstrip": false,
|
742 |
"normalized": false,
|
743 |
+
"rstrip": false,
|
744 |
"single_word": false,
|
745 |
"special": true
|
746 |
},
|
747 |
"93": {
|
748 |
"content": "<unk_91>",
|
749 |
+
"lstrip": false,
|
750 |
"normalized": false,
|
751 |
+
"rstrip": false,
|
752 |
"single_word": false,
|
753 |
"special": true
|
754 |
},
|
755 |
"94": {
|
756 |
"content": "<unk_92>",
|
757 |
+
"lstrip": false,
|
758 |
"normalized": false,
|
759 |
+
"rstrip": false,
|
760 |
"single_word": false,
|
761 |
"special": true
|
762 |
},
|
763 |
"95": {
|
764 |
"content": "<unk_93>",
|
765 |
+
"lstrip": false,
|
766 |
"normalized": false,
|
767 |
+
"rstrip": false,
|
768 |
"single_word": false,
|
769 |
"special": true
|
770 |
},
|
771 |
"96": {
|
772 |
"content": "<unk_94>",
|
773 |
+
"lstrip": false,
|
774 |
"normalized": false,
|
775 |
+
"rstrip": false,
|
776 |
"single_word": false,
|
777 |
"special": true
|
778 |
},
|
779 |
"97": {
|
780 |
"content": "<unk_95>",
|
781 |
+
"lstrip": false,
|
782 |
"normalized": false,
|
783 |
+
"rstrip": false,
|
784 |
"single_word": false,
|
785 |
"special": true
|
786 |
},
|
787 |
"98": {
|
788 |
"content": "<unk_96>",
|
789 |
+
"lstrip": false,
|
790 |
"normalized": false,
|
791 |
+
"rstrip": false,
|
792 |
"single_word": false,
|
793 |
"special": true
|
794 |
},
|
795 |
"99": {
|
796 |
"content": "<unk_97>",
|
797 |
+
"lstrip": false,
|
798 |
"normalized": false,
|
799 |
+
"rstrip": false,
|
800 |
"single_word": false,
|
801 |
"special": true
|
802 |
},
|
803 |
"100": {
|
804 |
"content": "<unk_98>",
|
805 |
+
"lstrip": false,
|
806 |
"normalized": false,
|
807 |
+
"rstrip": false,
|
808 |
"single_word": false,
|
809 |
"special": true
|
810 |
},
|
811 |
"101": {
|
812 |
"content": "<unk_99>",
|
813 |
+
"lstrip": false,
|
814 |
"normalized": false,
|
815 |
+
"rstrip": false,
|
816 |
"single_word": false,
|
817 |
"special": true
|
818 |
},
|
819 |
"102": {
|
820 |
"content": "<unk_100>",
|
821 |
+
"lstrip": false,
|
822 |
"normalized": false,
|
823 |
+
"rstrip": false,
|
824 |
"single_word": false,
|
825 |
"special": true
|
826 |
},
|
827 |
"103": {
|
828 |
"content": "<unk_101>",
|
829 |
+
"lstrip": false,
|
830 |
"normalized": false,
|
831 |
+
"rstrip": false,
|
832 |
"single_word": false,
|
833 |
"special": true
|
834 |
},
|
835 |
"104": {
|
836 |
"content": "<unk_102>",
|
837 |
+
"lstrip": false,
|
838 |
"normalized": false,
|
839 |
+
"rstrip": false,
|
840 |
"single_word": false,
|
841 |
"special": true
|
842 |
},
|
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 sha256:191b7fa152e554171c206f925284e85c4dc02518a33bc51111715c556cf141a0
|
3 |
+
size 4472
|