Commit
·
8ba0b1b
1
Parent(s):
b3fdbb8
Upload tokenizer
Browse files- special_tokens_map.json +7 -1
- tokenizer_config.json +8 -1
special_tokens_map.json
CHANGED
|
@@ -13,7 +13,13 @@
|
|
| 13 |
"rstrip": false,
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
-
"pad_token":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
"unk_token": {
|
| 18 |
"content": "<unk>",
|
| 19 |
"lstrip": false,
|
|
|
|
| 13 |
"rstrip": false,
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<|pad|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
"unk_token": {
|
| 24 |
"content": "<unk>",
|
| 25 |
"lstrip": false,
|
tokenizer_config.json
CHANGED
|
@@ -19,7 +19,14 @@
|
|
| 19 |
"single_word": false
|
| 20 |
},
|
| 21 |
"model_max_length": 1000000000000000019884624838656,
|
| 22 |
-
"pad_token":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
"sp_model_kwargs": {},
|
| 24 |
"tokenizer_class": "LlamaTokenizer",
|
| 25 |
"unk_token": {
|
|
|
|
| 19 |
"single_word": false
|
| 20 |
},
|
| 21 |
"model_max_length": 1000000000000000019884624838656,
|
| 22 |
+
"pad_token": {
|
| 23 |
+
"__type": "AddedToken",
|
| 24 |
+
"content": "<|pad|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
"sp_model_kwargs": {},
|
| 31 |
"tokenizer_class": "LlamaTokenizer",
|
| 32 |
"unk_token": {
|