Upload tokenizer
Browse files- tokenizer.json +2 -2
- tokenizer_config.json +1 -1
tokenizer.json
CHANGED
@@ -18,8 +18,8 @@
|
|
18 |
"single_word": false,
|
19 |
"lstrip": false,
|
20 |
"rstrip": false,
|
21 |
-
"normalized":
|
22 |
-
"special":
|
23 |
},
|
24 |
{
|
25 |
"id": 50001,
|
|
|
18 |
"single_word": false,
|
19 |
"lstrip": false,
|
20 |
"rstrip": false,
|
21 |
+
"normalized": true,
|
22 |
+
"special": false
|
23 |
},
|
24 |
{
|
25 |
"id": 50001,
|
tokenizer_config.json
CHANGED
@@ -3,7 +3,7 @@
|
|
3 |
"bos_token": "<|endoftext|>",
|
4 |
"clean_up_tokenization_spaces": true,
|
5 |
"eos_token": "<|endoftext|>",
|
6 |
-
"model_max_length":
|
7 |
"tokenizer_class": "GPT2Tokenizer",
|
8 |
"unk_token": "<|endoftext|>"
|
9 |
}
|
|
|
3 |
"bos_token": "<|endoftext|>",
|
4 |
"clean_up_tokenization_spaces": true,
|
5 |
"eos_token": "<|endoftext|>",
|
6 |
+
"model_max_length": 512,
|
7 |
"tokenizer_class": "GPT2Tokenizer",
|
8 |
"unk_token": "<|endoftext|>"
|
9 |
}
|