Upload tokenizer
Browse files- .gitattributes +1 -0
- tokenizer.json +3 -0
- tokenizer_config.json +7 -12
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1acc81dbe7b37b31f07eff79502d66f6fe37d9bc69ddb28a4befd818dbad1acd
|
| 3 |
+
size 31923943
|
tokenizer_config.json
CHANGED
|
@@ -1,25 +1,20 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
"bos_token": "<s>",
|
| 4 |
-
"clean_up_tokenization_spaces": true,
|
| 5 |
"cls_token": "<s>",
|
| 6 |
"eos_token": "</s>",
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
"legacy_behaviour": false,
|
| 8 |
-
"mask_token":
|
| 9 |
-
"__type": "AddedToken",
|
| 10 |
-
"content": "<mask>",
|
| 11 |
-
"lstrip": true,
|
| 12 |
-
"normalized": true,
|
| 13 |
-
"rstrip": false,
|
| 14 |
-
"single_word": false
|
| 15 |
-
},
|
| 16 |
"model_max_length": 1024,
|
| 17 |
"pad_token": "<pad>",
|
| 18 |
"sep_token": "</s>",
|
| 19 |
"sp_model_kwargs": {},
|
| 20 |
-
"src_lang":
|
| 21 |
"tgt_lang": null,
|
| 22 |
"tokenizer_class": "NllbTokenizer",
|
| 23 |
-
"tokenizer_file": null,
|
| 24 |
"unk_token": "<unk>"
|
| 25 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
"bos_token": "<s>",
|
|
|
|
| 4 |
"cls_token": "<s>",
|
| 5 |
"eos_token": "</s>",
|
| 6 |
+
"extra_special_tokens": [
|
| 7 |
+
"shi_Latn"
|
| 8 |
+
],
|
| 9 |
+
"is_local": false,
|
| 10 |
"legacy_behaviour": false,
|
| 11 |
+
"mask_token": "<mask>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
"model_max_length": 1024,
|
| 13 |
"pad_token": "<pad>",
|
| 14 |
"sep_token": "</s>",
|
| 15 |
"sp_model_kwargs": {},
|
| 16 |
+
"src_lang": "shi_Latn",
|
| 17 |
"tgt_lang": null,
|
| 18 |
"tokenizer_class": "NllbTokenizer",
|
|
|
|
| 19 |
"unk_token": "<unk>"
|
| 20 |
}
|