sho-takase commited on
Commit
b283e16
1 Parent(s): de3a413

add tokenizer_config

Browse files
Files changed (1) hide show
  1. tokenizer_config.json +16 -0
tokenizer_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "extra_ids": 0,
3
+ "do_lower_case": false,
4
+ "keep_accents": true,
5
+ "bos_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "unk_token": "<unk>",
8
+ "pad_token": "<pad>",
9
+ "mask_token": "<mask>",
10
+ "cls_token": "<cls>",
11
+ "sep_token": "<sep>",
12
+ "padding_side": "left",
13
+ "sp_model_kwargs": {},
14
+ "special_tokens_map_file": null,
15
+ "tokenizer_class": "T5Tokenizer"
16
+ }