Commit
•
65f6059
1
Parent(s):
7baf67e
Upload tokenizer
Browse files- special_tokens_map.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +9 -0
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "<cls>",
|
| 3 |
+
"mask_token": "<mask>",
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"sep_token": "<sep>"
|
| 6 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"clean_up_tokenization_spaces": true,
|
| 3 |
+
"cls_token": "<cls>",
|
| 4 |
+
"mask_token": "<mask>",
|
| 5 |
+
"model_max_length": 512,
|
| 6 |
+
"pad_token": "<pad>",
|
| 7 |
+
"sep_token": "<sep>",
|
| 8 |
+
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 9 |
+
}
|