Upload folder using huggingface_hub
Browse files- added_tokens.json +25 -25
- special_tokens_map.json +25 -25
- tokenizer.json +25 -25
- tokenizer_config.json +50 -50
added_tokens.json
CHANGED
@@ -7,37 +7,37 @@
|
|
7 |
"<B_USYS>": 151666,
|
8 |
"<C_A>": 151668,
|
9 |
"<C_Q>": 151667,
|
10 |
-
"<
|
11 |
-
"<
|
12 |
-
"<
|
13 |
-
"<
|
14 |
-
"<
|
15 |
-
"<
|
16 |
-
"<
|
17 |
-
"<
|
18 |
-
"<
|
19 |
-
"<
|
20 |
-
"<
|
21 |
-
"<
|
22 |
-
"<
|
23 |
"<calc_end>": 151674,
|
24 |
"<calc_start>": 151673,
|
25 |
"<function_calling>": 151672,
|
26 |
-
"<
|
27 |
-
"<
|
28 |
-
"<
|
29 |
-
"<
|
30 |
-
"<
|
31 |
"<inner_think>": 151675,
|
32 |
-
"<
|
33 |
-
"<
|
34 |
-
"<
|
35 |
-
"<
|
36 |
"<reserved_113>": 151692,
|
37 |
"<tool_call>": 151657,
|
38 |
-
"<
|
39 |
-
"<
|
40 |
-
"<
|
41 |
"<|box_end|>": 151649,
|
42 |
"<|box_start|>": 151648,
|
43 |
"<|endoftext|>": 151643,
|
|
|
7 |
"<B_USYS>": 151666,
|
8 |
"<C_A>": 151668,
|
9 |
"<C_Q>": 151667,
|
10 |
+
"<audio_delim_baichuan>": 151693,
|
11 |
+
"<audio_end_baichuan>": 151677,
|
12 |
+
"<audio_pad_baichuan>": 151678,
|
13 |
+
"<audio_start_baichuan>": 151676,
|
14 |
+
"<audiogen_end_baichuan>": 151701,
|
15 |
+
"<audiogen_start_baichuan>": 151700,
|
16 |
+
"<audiotext_end_baichuan>": 151698,
|
17 |
+
"<audiotext_pad_baichuan>": 151699,
|
18 |
+
"<audiotext_start_baichuan>": 151697,
|
19 |
+
"<baichuan_pad_token>": 151691,
|
20 |
+
"<box_delim_baichuan>": 151685,
|
21 |
+
"<box_end_baichuan>": 151684,
|
22 |
+
"<box_start_baichuan>": 151683,
|
23 |
"<calc_end>": 151674,
|
24 |
"<calc_start>": 151673,
|
25 |
"<function_calling>": 151672,
|
26 |
+
"<img_delim_baichuan>": 151688,
|
27 |
+
"<img_end_baichuan>": 151680,
|
28 |
+
"<img_newline_baichuan>": 151682,
|
29 |
+
"<img_pad_baichuan>": 151681,
|
30 |
+
"<img_start_baichuan>": 151679,
|
31 |
"<inner_think>": 151675,
|
32 |
+
"<polygon_end_baichuan>": 151690,
|
33 |
+
"<polygon_start_baichuan>": 151689,
|
34 |
+
"<ref_end_baichuan>": 151687,
|
35 |
+
"<ref_start_baichuan>": 151686,
|
36 |
"<reserved_113>": 151692,
|
37 |
"<tool_call>": 151657,
|
38 |
+
"<video_end_baichuan>": 151696,
|
39 |
+
"<video_palce_baichuan>": 151694,
|
40 |
+
"<video_start_baichuan>": 151695,
|
41 |
"<|box_end|>": 151649,
|
42 |
"<|box_start|>": 151648,
|
43 |
"<|endoftext|>": 151643,
|
special_tokens_map.json
CHANGED
@@ -24,32 +24,32 @@
|
|
24 |
"<calc_start>",
|
25 |
"<calc_end>",
|
26 |
"<inner_think>",
|
27 |
-
"<
|
28 |
-
"<
|
29 |
-
"<
|
30 |
-
"<
|
31 |
-
"<
|
32 |
-
"<
|
33 |
-
"<
|
34 |
-
"<
|
35 |
-
"<
|
36 |
-
"<
|
37 |
-
"<
|
38 |
-
"<
|
39 |
-
"<
|
40 |
-
"<
|
41 |
-
"<
|
42 |
-
"<
|
43 |
"<reserved_113>",
|
44 |
-
"<
|
45 |
-
"<
|
46 |
-
"<
|
47 |
-
"<
|
48 |
-
"<
|
49 |
-
"<
|
50 |
-
"<
|
51 |
-
"<
|
52 |
-
"<
|
53 |
],
|
54 |
"eos_token": {
|
55 |
"content": "<|endoftext|>",
|
|
|
24 |
"<calc_start>",
|
25 |
"<calc_end>",
|
26 |
"<inner_think>",
|
27 |
+
"<audio_start_baichuan>",
|
28 |
+
"<audio_end_baichuan>",
|
29 |
+
"<audio_pad_baichuan>",
|
30 |
+
"<img_start_baichuan>",
|
31 |
+
"<img_end_baichuan>",
|
32 |
+
"<img_pad_baichuan>",
|
33 |
+
"<img_newline_baichuan>",
|
34 |
+
"<box_start_baichuan>",
|
35 |
+
"<box_end_baichuan>",
|
36 |
+
"<box_delim_baichuan>",
|
37 |
+
"<ref_start_baichuan>",
|
38 |
+
"<ref_end_baichuan>",
|
39 |
+
"<img_delim_baichuan>",
|
40 |
+
"<polygon_start_baichuan>",
|
41 |
+
"<polygon_end_baichuan>",
|
42 |
+
"<baichuan_pad_token>",
|
43 |
"<reserved_113>",
|
44 |
+
"<audio_delim_baichuan>",
|
45 |
+
"<video_start_baichuan>",
|
46 |
+
"<video_end_baichuan>",
|
47 |
+
"<video_palce_baichuan>",
|
48 |
+
"<audiotext_start_baichuan>",
|
49 |
+
"<audiotext_end_baichuan>",
|
50 |
+
"<audiotext_pad_baichuan>",
|
51 |
+
"<audiogen_start_baichuan>",
|
52 |
+
"<audiogen_end_baichuan>"
|
53 |
],
|
54 |
"eos_token": {
|
55 |
"content": "<|endoftext|>",
|
tokenizer.json
CHANGED
@@ -307,7 +307,7 @@
|
|
307 |
},
|
308 |
{
|
309 |
"id": 151676,
|
310 |
-
"content": "<
|
311 |
"single_word": false,
|
312 |
"lstrip": false,
|
313 |
"rstrip": false,
|
@@ -316,7 +316,7 @@
|
|
316 |
},
|
317 |
{
|
318 |
"id": 151677,
|
319 |
-
"content": "<
|
320 |
"single_word": false,
|
321 |
"lstrip": false,
|
322 |
"rstrip": false,
|
@@ -325,7 +325,7 @@
|
|
325 |
},
|
326 |
{
|
327 |
"id": 151678,
|
328 |
-
"content": "<
|
329 |
"single_word": false,
|
330 |
"lstrip": false,
|
331 |
"rstrip": false,
|
@@ -334,7 +334,7 @@
|
|
334 |
},
|
335 |
{
|
336 |
"id": 151679,
|
337 |
-
"content": "<
|
338 |
"single_word": false,
|
339 |
"lstrip": false,
|
340 |
"rstrip": false,
|
@@ -343,7 +343,7 @@
|
|
343 |
},
|
344 |
{
|
345 |
"id": 151680,
|
346 |
-
"content": "<
|
347 |
"single_word": false,
|
348 |
"lstrip": false,
|
349 |
"rstrip": false,
|
@@ -352,7 +352,7 @@
|
|
352 |
},
|
353 |
{
|
354 |
"id": 151681,
|
355 |
-
"content": "<
|
356 |
"single_word": false,
|
357 |
"lstrip": false,
|
358 |
"rstrip": false,
|
@@ -361,7 +361,7 @@
|
|
361 |
},
|
362 |
{
|
363 |
"id": 151682,
|
364 |
-
"content": "<
|
365 |
"single_word": false,
|
366 |
"lstrip": false,
|
367 |
"rstrip": false,
|
@@ -370,7 +370,7 @@
|
|
370 |
},
|
371 |
{
|
372 |
"id": 151683,
|
373 |
-
"content": "<
|
374 |
"single_word": false,
|
375 |
"lstrip": false,
|
376 |
"rstrip": false,
|
@@ -379,7 +379,7 @@
|
|
379 |
},
|
380 |
{
|
381 |
"id": 151684,
|
382 |
-
"content": "<
|
383 |
"single_word": false,
|
384 |
"lstrip": false,
|
385 |
"rstrip": false,
|
@@ -388,7 +388,7 @@
|
|
388 |
},
|
389 |
{
|
390 |
"id": 151685,
|
391 |
-
"content": "<
|
392 |
"single_word": false,
|
393 |
"lstrip": false,
|
394 |
"rstrip": false,
|
@@ -397,7 +397,7 @@
|
|
397 |
},
|
398 |
{
|
399 |
"id": 151686,
|
400 |
-
"content": "<
|
401 |
"single_word": false,
|
402 |
"lstrip": false,
|
403 |
"rstrip": false,
|
@@ -406,7 +406,7 @@
|
|
406 |
},
|
407 |
{
|
408 |
"id": 151687,
|
409 |
-
"content": "<
|
410 |
"single_word": false,
|
411 |
"lstrip": false,
|
412 |
"rstrip": false,
|
@@ -415,7 +415,7 @@
|
|
415 |
},
|
416 |
{
|
417 |
"id": 151688,
|
418 |
-
"content": "<
|
419 |
"single_word": false,
|
420 |
"lstrip": false,
|
421 |
"rstrip": false,
|
@@ -424,7 +424,7 @@
|
|
424 |
},
|
425 |
{
|
426 |
"id": 151689,
|
427 |
-
"content": "<
|
428 |
"single_word": false,
|
429 |
"lstrip": false,
|
430 |
"rstrip": false,
|
@@ -433,7 +433,7 @@
|
|
433 |
},
|
434 |
{
|
435 |
"id": 151690,
|
436 |
-
"content": "<
|
437 |
"single_word": false,
|
438 |
"lstrip": false,
|
439 |
"rstrip": false,
|
@@ -442,7 +442,7 @@
|
|
442 |
},
|
443 |
{
|
444 |
"id": 151691,
|
445 |
-
"content": "<
|
446 |
"single_word": false,
|
447 |
"lstrip": false,
|
448 |
"rstrip": false,
|
@@ -460,7 +460,7 @@
|
|
460 |
},
|
461 |
{
|
462 |
"id": 151693,
|
463 |
-
"content": "<
|
464 |
"single_word": false,
|
465 |
"lstrip": false,
|
466 |
"rstrip": false,
|
@@ -469,7 +469,7 @@
|
|
469 |
},
|
470 |
{
|
471 |
"id": 151694,
|
472 |
-
"content": "<
|
473 |
"single_word": false,
|
474 |
"lstrip": false,
|
475 |
"rstrip": false,
|
@@ -478,7 +478,7 @@
|
|
478 |
},
|
479 |
{
|
480 |
"id": 151695,
|
481 |
-
"content": "<
|
482 |
"single_word": false,
|
483 |
"lstrip": false,
|
484 |
"rstrip": false,
|
@@ -487,7 +487,7 @@
|
|
487 |
},
|
488 |
{
|
489 |
"id": 151696,
|
490 |
-
"content": "<
|
491 |
"single_word": false,
|
492 |
"lstrip": false,
|
493 |
"rstrip": false,
|
@@ -496,7 +496,7 @@
|
|
496 |
},
|
497 |
{
|
498 |
"id": 151697,
|
499 |
-
"content": "<
|
500 |
"single_word": false,
|
501 |
"lstrip": false,
|
502 |
"rstrip": false,
|
@@ -505,7 +505,7 @@
|
|
505 |
},
|
506 |
{
|
507 |
"id": 151698,
|
508 |
-
"content": "<
|
509 |
"single_word": false,
|
510 |
"lstrip": false,
|
511 |
"rstrip": false,
|
@@ -514,7 +514,7 @@
|
|
514 |
},
|
515 |
{
|
516 |
"id": 151699,
|
517 |
-
"content": "<
|
518 |
"single_word": false,
|
519 |
"lstrip": false,
|
520 |
"rstrip": false,
|
@@ -523,7 +523,7 @@
|
|
523 |
},
|
524 |
{
|
525 |
"id": 151700,
|
526 |
-
"content": "<
|
527 |
"single_word": false,
|
528 |
"lstrip": false,
|
529 |
"rstrip": false,
|
@@ -532,7 +532,7 @@
|
|
532 |
},
|
533 |
{
|
534 |
"id": 151701,
|
535 |
-
"content": "<
|
536 |
"single_word": false,
|
537 |
"lstrip": false,
|
538 |
"rstrip": false,
|
|
|
307 |
},
|
308 |
{
|
309 |
"id": 151676,
|
310 |
+
"content": "<audio_start_baichuan>",
|
311 |
"single_word": false,
|
312 |
"lstrip": false,
|
313 |
"rstrip": false,
|
|
|
316 |
},
|
317 |
{
|
318 |
"id": 151677,
|
319 |
+
"content": "<audio_end_baichuan>",
|
320 |
"single_word": false,
|
321 |
"lstrip": false,
|
322 |
"rstrip": false,
|
|
|
325 |
},
|
326 |
{
|
327 |
"id": 151678,
|
328 |
+
"content": "<audio_pad_baichuan>",
|
329 |
"single_word": false,
|
330 |
"lstrip": false,
|
331 |
"rstrip": false,
|
|
|
334 |
},
|
335 |
{
|
336 |
"id": 151679,
|
337 |
+
"content": "<img_start_baichuan>",
|
338 |
"single_word": false,
|
339 |
"lstrip": false,
|
340 |
"rstrip": false,
|
|
|
343 |
},
|
344 |
{
|
345 |
"id": 151680,
|
346 |
+
"content": "<img_end_baichuan>",
|
347 |
"single_word": false,
|
348 |
"lstrip": false,
|
349 |
"rstrip": false,
|
|
|
352 |
},
|
353 |
{
|
354 |
"id": 151681,
|
355 |
+
"content": "<img_pad_baichuan>",
|
356 |
"single_word": false,
|
357 |
"lstrip": false,
|
358 |
"rstrip": false,
|
|
|
361 |
},
|
362 |
{
|
363 |
"id": 151682,
|
364 |
+
"content": "<img_newline_baichuan>",
|
365 |
"single_word": false,
|
366 |
"lstrip": false,
|
367 |
"rstrip": false,
|
|
|
370 |
},
|
371 |
{
|
372 |
"id": 151683,
|
373 |
+
"content": "<box_start_baichuan>",
|
374 |
"single_word": false,
|
375 |
"lstrip": false,
|
376 |
"rstrip": false,
|
|
|
379 |
},
|
380 |
{
|
381 |
"id": 151684,
|
382 |
+
"content": "<box_end_baichuan>",
|
383 |
"single_word": false,
|
384 |
"lstrip": false,
|
385 |
"rstrip": false,
|
|
|
388 |
},
|
389 |
{
|
390 |
"id": 151685,
|
391 |
+
"content": "<box_delim_baichuan>",
|
392 |
"single_word": false,
|
393 |
"lstrip": false,
|
394 |
"rstrip": false,
|
|
|
397 |
},
|
398 |
{
|
399 |
"id": 151686,
|
400 |
+
"content": "<ref_start_baichuan>",
|
401 |
"single_word": false,
|
402 |
"lstrip": false,
|
403 |
"rstrip": false,
|
|
|
406 |
},
|
407 |
{
|
408 |
"id": 151687,
|
409 |
+
"content": "<ref_end_baichuan>",
|
410 |
"single_word": false,
|
411 |
"lstrip": false,
|
412 |
"rstrip": false,
|
|
|
415 |
},
|
416 |
{
|
417 |
"id": 151688,
|
418 |
+
"content": "<img_delim_baichuan>",
|
419 |
"single_word": false,
|
420 |
"lstrip": false,
|
421 |
"rstrip": false,
|
|
|
424 |
},
|
425 |
{
|
426 |
"id": 151689,
|
427 |
+
"content": "<polygon_start_baichuan>",
|
428 |
"single_word": false,
|
429 |
"lstrip": false,
|
430 |
"rstrip": false,
|
|
|
433 |
},
|
434 |
{
|
435 |
"id": 151690,
|
436 |
+
"content": "<polygon_end_baichuan>",
|
437 |
"single_word": false,
|
438 |
"lstrip": false,
|
439 |
"rstrip": false,
|
|
|
442 |
},
|
443 |
{
|
444 |
"id": 151691,
|
445 |
+
"content": "<baichuan_pad_token>",
|
446 |
"single_word": false,
|
447 |
"lstrip": false,
|
448 |
"rstrip": false,
|
|
|
460 |
},
|
461 |
{
|
462 |
"id": 151693,
|
463 |
+
"content": "<audio_delim_baichuan>",
|
464 |
"single_word": false,
|
465 |
"lstrip": false,
|
466 |
"rstrip": false,
|
|
|
469 |
},
|
470 |
{
|
471 |
"id": 151694,
|
472 |
+
"content": "<video_palce_baichuan>",
|
473 |
"single_word": false,
|
474 |
"lstrip": false,
|
475 |
"rstrip": false,
|
|
|
478 |
},
|
479 |
{
|
480 |
"id": 151695,
|
481 |
+
"content": "<video_start_baichuan>",
|
482 |
"single_word": false,
|
483 |
"lstrip": false,
|
484 |
"rstrip": false,
|
|
|
487 |
},
|
488 |
{
|
489 |
"id": 151696,
|
490 |
+
"content": "<video_end_baichuan>",
|
491 |
"single_word": false,
|
492 |
"lstrip": false,
|
493 |
"rstrip": false,
|
|
|
496 |
},
|
497 |
{
|
498 |
"id": 151697,
|
499 |
+
"content": "<audiotext_start_baichuan>",
|
500 |
"single_word": false,
|
501 |
"lstrip": false,
|
502 |
"rstrip": false,
|
|
|
505 |
},
|
506 |
{
|
507 |
"id": 151698,
|
508 |
+
"content": "<audiotext_end_baichuan>",
|
509 |
"single_word": false,
|
510 |
"lstrip": false,
|
511 |
"rstrip": false,
|
|
|
514 |
},
|
515 |
{
|
516 |
"id": 151699,
|
517 |
+
"content": "<audiotext_pad_baichuan>",
|
518 |
"single_word": false,
|
519 |
"lstrip": false,
|
520 |
"rstrip": false,
|
|
|
523 |
},
|
524 |
{
|
525 |
"id": 151700,
|
526 |
+
"content": "<audiogen_start_baichuan>",
|
527 |
"single_word": false,
|
528 |
"lstrip": false,
|
529 |
"rstrip": false,
|
|
|
532 |
},
|
533 |
{
|
534 |
"id": 151701,
|
535 |
+
"content": "<audiogen_end_baichuan>",
|
536 |
"single_word": false,
|
537 |
"lstrip": false,
|
538 |
"rstrip": false,
|
tokenizer_config.json
CHANGED
@@ -267,7 +267,7 @@
|
|
267 |
"special": true
|
268 |
},
|
269 |
"151676": {
|
270 |
-
"content": "<
|
271 |
"lstrip": false,
|
272 |
"normalized": false,
|
273 |
"rstrip": false,
|
@@ -275,7 +275,7 @@
|
|
275 |
"special": true
|
276 |
},
|
277 |
"151677": {
|
278 |
-
"content": "<
|
279 |
"lstrip": false,
|
280 |
"normalized": false,
|
281 |
"rstrip": false,
|
@@ -283,7 +283,7 @@
|
|
283 |
"special": true
|
284 |
},
|
285 |
"151678": {
|
286 |
-
"content": "<
|
287 |
"lstrip": false,
|
288 |
"normalized": false,
|
289 |
"rstrip": false,
|
@@ -291,7 +291,7 @@
|
|
291 |
"special": true
|
292 |
},
|
293 |
"151679": {
|
294 |
-
"content": "<
|
295 |
"lstrip": false,
|
296 |
"normalized": false,
|
297 |
"rstrip": false,
|
@@ -299,7 +299,7 @@
|
|
299 |
"special": true
|
300 |
},
|
301 |
"151680": {
|
302 |
-
"content": "<
|
303 |
"lstrip": false,
|
304 |
"normalized": false,
|
305 |
"rstrip": false,
|
@@ -307,7 +307,7 @@
|
|
307 |
"special": true
|
308 |
},
|
309 |
"151681": {
|
310 |
-
"content": "<
|
311 |
"lstrip": false,
|
312 |
"normalized": false,
|
313 |
"rstrip": false,
|
@@ -315,7 +315,7 @@
|
|
315 |
"special": true
|
316 |
},
|
317 |
"151682": {
|
318 |
-
"content": "<
|
319 |
"lstrip": false,
|
320 |
"normalized": false,
|
321 |
"rstrip": false,
|
@@ -323,7 +323,7 @@
|
|
323 |
"special": true
|
324 |
},
|
325 |
"151683": {
|
326 |
-
"content": "<
|
327 |
"lstrip": false,
|
328 |
"normalized": false,
|
329 |
"rstrip": false,
|
@@ -331,7 +331,7 @@
|
|
331 |
"special": true
|
332 |
},
|
333 |
"151684": {
|
334 |
-
"content": "<
|
335 |
"lstrip": false,
|
336 |
"normalized": false,
|
337 |
"rstrip": false,
|
@@ -339,7 +339,7 @@
|
|
339 |
"special": true
|
340 |
},
|
341 |
"151685": {
|
342 |
-
"content": "<
|
343 |
"lstrip": false,
|
344 |
"normalized": false,
|
345 |
"rstrip": false,
|
@@ -347,7 +347,7 @@
|
|
347 |
"special": true
|
348 |
},
|
349 |
"151686": {
|
350 |
-
"content": "<
|
351 |
"lstrip": false,
|
352 |
"normalized": false,
|
353 |
"rstrip": false,
|
@@ -355,7 +355,7 @@
|
|
355 |
"special": true
|
356 |
},
|
357 |
"151687": {
|
358 |
-
"content": "<
|
359 |
"lstrip": false,
|
360 |
"normalized": false,
|
361 |
"rstrip": false,
|
@@ -363,7 +363,7 @@
|
|
363 |
"special": true
|
364 |
},
|
365 |
"151688": {
|
366 |
-
"content": "<
|
367 |
"lstrip": false,
|
368 |
"normalized": false,
|
369 |
"rstrip": false,
|
@@ -371,7 +371,7 @@
|
|
371 |
"special": true
|
372 |
},
|
373 |
"151689": {
|
374 |
-
"content": "<
|
375 |
"lstrip": false,
|
376 |
"normalized": false,
|
377 |
"rstrip": false,
|
@@ -379,7 +379,7 @@
|
|
379 |
"special": true
|
380 |
},
|
381 |
"151690": {
|
382 |
-
"content": "<
|
383 |
"lstrip": false,
|
384 |
"normalized": false,
|
385 |
"rstrip": false,
|
@@ -387,7 +387,7 @@
|
|
387 |
"special": true
|
388 |
},
|
389 |
"151691": {
|
390 |
-
"content": "<
|
391 |
"lstrip": false,
|
392 |
"normalized": false,
|
393 |
"rstrip": false,
|
@@ -403,7 +403,7 @@
|
|
403 |
"special": true
|
404 |
},
|
405 |
"151693": {
|
406 |
-
"content": "<
|
407 |
"lstrip": false,
|
408 |
"normalized": false,
|
409 |
"rstrip": false,
|
@@ -411,7 +411,7 @@
|
|
411 |
"special": true
|
412 |
},
|
413 |
"151694": {
|
414 |
-
"content": "<
|
415 |
"lstrip": false,
|
416 |
"normalized": false,
|
417 |
"rstrip": false,
|
@@ -419,7 +419,7 @@
|
|
419 |
"special": true
|
420 |
},
|
421 |
"151695": {
|
422 |
-
"content": "<
|
423 |
"lstrip": false,
|
424 |
"normalized": false,
|
425 |
"rstrip": false,
|
@@ -427,7 +427,7 @@
|
|
427 |
"special": true
|
428 |
},
|
429 |
"151696": {
|
430 |
-
"content": "<
|
431 |
"lstrip": false,
|
432 |
"normalized": false,
|
433 |
"rstrip": false,
|
@@ -435,7 +435,7 @@
|
|
435 |
"special": true
|
436 |
},
|
437 |
"151697": {
|
438 |
-
"content": "<
|
439 |
"lstrip": false,
|
440 |
"normalized": false,
|
441 |
"rstrip": false,
|
@@ -443,7 +443,7 @@
|
|
443 |
"special": true
|
444 |
},
|
445 |
"151698": {
|
446 |
-
"content": "<
|
447 |
"lstrip": false,
|
448 |
"normalized": false,
|
449 |
"rstrip": false,
|
@@ -451,7 +451,7 @@
|
|
451 |
"special": true
|
452 |
},
|
453 |
"151699": {
|
454 |
-
"content": "<
|
455 |
"lstrip": false,
|
456 |
"normalized": false,
|
457 |
"rstrip": false,
|
@@ -459,7 +459,7 @@
|
|
459 |
"special": true
|
460 |
},
|
461 |
"151700": {
|
462 |
-
"content": "<
|
463 |
"lstrip": false,
|
464 |
"normalized": false,
|
465 |
"rstrip": false,
|
@@ -467,7 +467,7 @@
|
|
467 |
"special": true
|
468 |
},
|
469 |
"151701": {
|
470 |
-
"content": "<
|
471 |
"lstrip": false,
|
472 |
"normalized": false,
|
473 |
"rstrip": false,
|
@@ -500,32 +500,32 @@
|
|
500 |
"<calc_start>",
|
501 |
"<calc_end>",
|
502 |
"<inner_think>",
|
503 |
-
"<
|
504 |
-
"<
|
505 |
-
"<
|
506 |
-
"<
|
507 |
-
"<
|
508 |
-
"<
|
509 |
-
"<
|
510 |
-
"<
|
511 |
-
"<
|
512 |
-
"<
|
513 |
-
"<
|
514 |
-
"<
|
515 |
-
"<
|
516 |
-
"<
|
517 |
-
"<
|
518 |
-
"<
|
519 |
"<reserved_113>",
|
520 |
-
"<
|
521 |
-
"<
|
522 |
-
"<
|
523 |
-
"<
|
524 |
-
"<
|
525 |
-
"<
|
526 |
-
"<
|
527 |
-
"<
|
528 |
-
"<
|
529 |
],
|
530 |
"bos_token": null,
|
531 |
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|
|
|
267 |
"special": true
|
268 |
},
|
269 |
"151676": {
|
270 |
+
"content": "<audio_start_baichuan>",
|
271 |
"lstrip": false,
|
272 |
"normalized": false,
|
273 |
"rstrip": false,
|
|
|
275 |
"special": true
|
276 |
},
|
277 |
"151677": {
|
278 |
+
"content": "<audio_end_baichuan>",
|
279 |
"lstrip": false,
|
280 |
"normalized": false,
|
281 |
"rstrip": false,
|
|
|
283 |
"special": true
|
284 |
},
|
285 |
"151678": {
|
286 |
+
"content": "<audio_pad_baichuan>",
|
287 |
"lstrip": false,
|
288 |
"normalized": false,
|
289 |
"rstrip": false,
|
|
|
291 |
"special": true
|
292 |
},
|
293 |
"151679": {
|
294 |
+
"content": "<img_start_baichuan>",
|
295 |
"lstrip": false,
|
296 |
"normalized": false,
|
297 |
"rstrip": false,
|
|
|
299 |
"special": true
|
300 |
},
|
301 |
"151680": {
|
302 |
+
"content": "<img_end_baichuan>",
|
303 |
"lstrip": false,
|
304 |
"normalized": false,
|
305 |
"rstrip": false,
|
|
|
307 |
"special": true
|
308 |
},
|
309 |
"151681": {
|
310 |
+
"content": "<img_pad_baichuan>",
|
311 |
"lstrip": false,
|
312 |
"normalized": false,
|
313 |
"rstrip": false,
|
|
|
315 |
"special": true
|
316 |
},
|
317 |
"151682": {
|
318 |
+
"content": "<img_newline_baichuan>",
|
319 |
"lstrip": false,
|
320 |
"normalized": false,
|
321 |
"rstrip": false,
|
|
|
323 |
"special": true
|
324 |
},
|
325 |
"151683": {
|
326 |
+
"content": "<box_start_baichuan>",
|
327 |
"lstrip": false,
|
328 |
"normalized": false,
|
329 |
"rstrip": false,
|
|
|
331 |
"special": true
|
332 |
},
|
333 |
"151684": {
|
334 |
+
"content": "<box_end_baichuan>",
|
335 |
"lstrip": false,
|
336 |
"normalized": false,
|
337 |
"rstrip": false,
|
|
|
339 |
"special": true
|
340 |
},
|
341 |
"151685": {
|
342 |
+
"content": "<box_delim_baichuan>",
|
343 |
"lstrip": false,
|
344 |
"normalized": false,
|
345 |
"rstrip": false,
|
|
|
347 |
"special": true
|
348 |
},
|
349 |
"151686": {
|
350 |
+
"content": "<ref_start_baichuan>",
|
351 |
"lstrip": false,
|
352 |
"normalized": false,
|
353 |
"rstrip": false,
|
|
|
355 |
"special": true
|
356 |
},
|
357 |
"151687": {
|
358 |
+
"content": "<ref_end_baichuan>",
|
359 |
"lstrip": false,
|
360 |
"normalized": false,
|
361 |
"rstrip": false,
|
|
|
363 |
"special": true
|
364 |
},
|
365 |
"151688": {
|
366 |
+
"content": "<img_delim_baichuan>",
|
367 |
"lstrip": false,
|
368 |
"normalized": false,
|
369 |
"rstrip": false,
|
|
|
371 |
"special": true
|
372 |
},
|
373 |
"151689": {
|
374 |
+
"content": "<polygon_start_baichuan>",
|
375 |
"lstrip": false,
|
376 |
"normalized": false,
|
377 |
"rstrip": false,
|
|
|
379 |
"special": true
|
380 |
},
|
381 |
"151690": {
|
382 |
+
"content": "<polygon_end_baichuan>",
|
383 |
"lstrip": false,
|
384 |
"normalized": false,
|
385 |
"rstrip": false,
|
|
|
387 |
"special": true
|
388 |
},
|
389 |
"151691": {
|
390 |
+
"content": "<baichuan_pad_token>",
|
391 |
"lstrip": false,
|
392 |
"normalized": false,
|
393 |
"rstrip": false,
|
|
|
403 |
"special": true
|
404 |
},
|
405 |
"151693": {
|
406 |
+
"content": "<audio_delim_baichuan>",
|
407 |
"lstrip": false,
|
408 |
"normalized": false,
|
409 |
"rstrip": false,
|
|
|
411 |
"special": true
|
412 |
},
|
413 |
"151694": {
|
414 |
+
"content": "<video_palce_baichuan>",
|
415 |
"lstrip": false,
|
416 |
"normalized": false,
|
417 |
"rstrip": false,
|
|
|
419 |
"special": true
|
420 |
},
|
421 |
"151695": {
|
422 |
+
"content": "<video_start_baichuan>",
|
423 |
"lstrip": false,
|
424 |
"normalized": false,
|
425 |
"rstrip": false,
|
|
|
427 |
"special": true
|
428 |
},
|
429 |
"151696": {
|
430 |
+
"content": "<video_end_baichuan>",
|
431 |
"lstrip": false,
|
432 |
"normalized": false,
|
433 |
"rstrip": false,
|
|
|
435 |
"special": true
|
436 |
},
|
437 |
"151697": {
|
438 |
+
"content": "<audiotext_start_baichuan>",
|
439 |
"lstrip": false,
|
440 |
"normalized": false,
|
441 |
"rstrip": false,
|
|
|
443 |
"special": true
|
444 |
},
|
445 |
"151698": {
|
446 |
+
"content": "<audiotext_end_baichuan>",
|
447 |
"lstrip": false,
|
448 |
"normalized": false,
|
449 |
"rstrip": false,
|
|
|
451 |
"special": true
|
452 |
},
|
453 |
"151699": {
|
454 |
+
"content": "<audiotext_pad_baichuan>",
|
455 |
"lstrip": false,
|
456 |
"normalized": false,
|
457 |
"rstrip": false,
|
|
|
459 |
"special": true
|
460 |
},
|
461 |
"151700": {
|
462 |
+
"content": "<audiogen_start_baichuan>",
|
463 |
"lstrip": false,
|
464 |
"normalized": false,
|
465 |
"rstrip": false,
|
|
|
467 |
"special": true
|
468 |
},
|
469 |
"151701": {
|
470 |
+
"content": "<audiogen_end_baichuan>",
|
471 |
"lstrip": false,
|
472 |
"normalized": false,
|
473 |
"rstrip": false,
|
|
|
500 |
"<calc_start>",
|
501 |
"<calc_end>",
|
502 |
"<inner_think>",
|
503 |
+
"<audio_start_baichuan>",
|
504 |
+
"<audio_end_baichuan>",
|
505 |
+
"<audio_pad_baichuan>",
|
506 |
+
"<img_start_baichuan>",
|
507 |
+
"<img_end_baichuan>",
|
508 |
+
"<img_pad_baichuan>",
|
509 |
+
"<img_newline_baichuan>",
|
510 |
+
"<box_start_baichuan>",
|
511 |
+
"<box_end_baichuan>",
|
512 |
+
"<box_delim_baichuan>",
|
513 |
+
"<ref_start_baichuan>",
|
514 |
+
"<ref_end_baichuan>",
|
515 |
+
"<img_delim_baichuan>",
|
516 |
+
"<polygon_start_baichuan>",
|
517 |
+
"<polygon_end_baichuan>",
|
518 |
+
"<baichuan_pad_token>",
|
519 |
"<reserved_113>",
|
520 |
+
"<audio_delim_baichuan>",
|
521 |
+
"<video_start_baichuan>",
|
522 |
+
"<video_end_baichuan>",
|
523 |
+
"<video_palce_baichuan>",
|
524 |
+
"<audiotext_start_baichuan>",
|
525 |
+
"<audiotext_end_baichuan>",
|
526 |
+
"<audiotext_pad_baichuan>",
|
527 |
+
"<audiogen_start_baichuan>",
|
528 |
+
"<audiogen_end_baichuan>"
|
529 |
],
|
530 |
"bos_token": null,
|
531 |
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|