Upload tokenizer
Browse files- special_tokens_map.json +0 -9
- tokenizer.json +0 -9
- tokenizer_config.json +1 -12
special_tokens_map.json
CHANGED
|
@@ -1,13 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"additional_special_tokens": [
|
| 3 |
-
{
|
| 4 |
-
"content": "<N>",
|
| 5 |
-
"lstrip": false,
|
| 6 |
-
"normalized": false,
|
| 7 |
-
"rstrip": false,
|
| 8 |
-
"single_word": false
|
| 9 |
-
}
|
| 10 |
-
],
|
| 11 |
"cls_token": "[CLS]",
|
| 12 |
"mask_token": "[MASK]",
|
| 13 |
"pad_token": "[PAD]",
|
|
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
"cls_token": "[CLS]",
|
| 3 |
"mask_token": "[MASK]",
|
| 4 |
"pad_token": "[PAD]",
|
tokenizer.json
CHANGED
|
@@ -47,15 +47,6 @@
|
|
| 47 |
"rstrip": false,
|
| 48 |
"normalized": false,
|
| 49 |
"special": true
|
| 50 |
-
},
|
| 51 |
-
{
|
| 52 |
-
"id": 28996,
|
| 53 |
-
"content": "<N>",
|
| 54 |
-
"single_word": false,
|
| 55 |
-
"lstrip": false,
|
| 56 |
-
"rstrip": false,
|
| 57 |
-
"normalized": false,
|
| 58 |
-
"special": true
|
| 59 |
}
|
| 60 |
],
|
| 61 |
"normalizer": {
|
|
|
|
| 47 |
"rstrip": false,
|
| 48 |
"normalized": false,
|
| 49 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
}
|
| 51 |
],
|
| 52 |
"normalizer": {
|
tokenizer_config.json
CHANGED
|
@@ -39,24 +39,13 @@
|
|
| 39 |
"rstrip": false,
|
| 40 |
"single_word": false,
|
| 41 |
"special": true
|
| 42 |
-
},
|
| 43 |
-
"28996": {
|
| 44 |
-
"content": "<N>",
|
| 45 |
-
"lstrip": false,
|
| 46 |
-
"normalized": false,
|
| 47 |
-
"rstrip": false,
|
| 48 |
-
"single_word": false,
|
| 49 |
-
"special": true
|
| 50 |
}
|
| 51 |
},
|
| 52 |
-
"additional_special_tokens": [
|
| 53 |
-
"<N>"
|
| 54 |
-
],
|
| 55 |
"clean_up_tokenization_spaces": true,
|
| 56 |
"cls_token": "[CLS]",
|
| 57 |
"do_lower_case": false,
|
| 58 |
"mask_token": "[MASK]",
|
| 59 |
-
"model_max_length":
|
| 60 |
"pad_token": "[PAD]",
|
| 61 |
"sep_token": "[SEP]",
|
| 62 |
"strip_accents": null,
|
|
|
|
| 39 |
"rstrip": false,
|
| 40 |
"single_word": false,
|
| 41 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
}
|
| 43 |
},
|
|
|
|
|
|
|
|
|
|
| 44 |
"clean_up_tokenization_spaces": true,
|
| 45 |
"cls_token": "[CLS]",
|
| 46 |
"do_lower_case": false,
|
| 47 |
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 256,
|
| 49 |
"pad_token": "[PAD]",
|
| 50 |
"sep_token": "[SEP]",
|
| 51 |
"strip_accents": null,
|