Upload tokenizer
Browse files- special_tokens_map.json +3 -1
- tokenizer.json +21 -1
- tokenizer_config.json +18 -0
special_tokens_map.json
CHANGED
|
@@ -1,4 +1,6 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"mask_token": "[MASK]",
|
| 3 |
-
"pad_token": "PADDING_TOKEN"
|
|
|
|
| 4 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "PADDING_TOKEN",
|
| 5 |
+
"sep_token": "[SEP]"
|
| 6 |
}
|
tokenizer.json
CHANGED
|
@@ -14,6 +14,24 @@
|
|
| 14 |
},
|
| 15 |
{
|
| 16 |
"id": 400006,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
"content": "[MASK]",
|
| 18 |
"single_word": false,
|
| 19 |
"lstrip": false,
|
|
@@ -400036,7 +400054,9 @@
|
|
| 400036 |
"[unused0]": 400002,
|
| 400037 |
"[unused1]": 400003,
|
| 400038 |
"[Q]": 400004,
|
| 400039 |
-
"[D]": 400005
|
|
|
|
|
|
|
| 400040 |
},
|
| 400041 |
"unk_token": "[UNK]"
|
| 400042 |
}
|
|
|
|
| 14 |
},
|
| 15 |
{
|
| 16 |
"id": 400006,
|
| 17 |
+
"content": "[CLS]",
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"special": true
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"id": 400007,
|
| 26 |
+
"content": "[SEP]",
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"special": true
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"id": 400008,
|
| 35 |
"content": "[MASK]",
|
| 36 |
"single_word": false,
|
| 37 |
"lstrip": false,
|
|
|
|
| 400054 |
"[unused0]": 400002,
|
| 400055 |
"[unused1]": 400003,
|
| 400056 |
"[Q]": 400004,
|
| 400057 |
+
"[D]": 400005,
|
| 400058 |
+
"[CLS]": 400006,
|
| 400059 |
+
"[SEP]": 400007
|
| 400060 |
},
|
| 400061 |
"unk_token": "[UNK]"
|
| 400062 |
}
|
tokenizer_config.json
CHANGED
|
@@ -9,6 +9,22 @@
|
|
| 9 |
"special": true
|
| 10 |
},
|
| 11 |
"400006": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
"content": "[MASK]",
|
| 13 |
"lstrip": false,
|
| 14 |
"normalized": false,
|
|
@@ -18,8 +34,10 @@
|
|
| 18 |
}
|
| 19 |
},
|
| 20 |
"clean_up_tokenization_spaces": false,
|
|
|
|
| 21 |
"mask_token": "[MASK]",
|
| 22 |
"model_max_length": 1000000000000000019884624838656,
|
| 23 |
"pad_token": "PADDING_TOKEN",
|
|
|
|
| 24 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 25 |
}
|
|
|
|
| 9 |
"special": true
|
| 10 |
},
|
| 11 |
"400006": {
|
| 12 |
+
"content": "[CLS]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"400007": {
|
| 20 |
+
"content": "[SEP]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"400008": {
|
| 28 |
"content": "[MASK]",
|
| 29 |
"lstrip": false,
|
| 30 |
"normalized": false,
|
|
|
|
| 34 |
}
|
| 35 |
},
|
| 36 |
"clean_up_tokenization_spaces": false,
|
| 37 |
+
"cls_token": "[CLS]",
|
| 38 |
"mask_token": "[MASK]",
|
| 39 |
"model_max_length": 1000000000000000019884624838656,
|
| 40 |
"pad_token": "PADDING_TOKEN",
|
| 41 |
+
"sep_token": "[SEP]",
|
| 42 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 43 |
}
|