caprizone6 commited on
Commit
0fce93e
1 Parent(s): 9cd9890

Upload tokenizer

Browse files
Files changed (4) hide show
  1. merges.txt +0 -0
  2. special_tokens_map.json +7 -0
  3. tokenizer_config.json +10 -0
  4. vocab.json +0 -0
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "<pad>",
5
+ "sep_token": "</s>",
6
+ "unk_token": "<unk>"
7
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "clean_up_tokenization_spaces": true,
4
+ "eos_token": "</s>",
5
+ "model_max_length": 1024,
6
+ "pad_token": "<pad>",
7
+ "sep_token": "</s>",
8
+ "tokenizer_class": "BioGptTokenizer",
9
+ "unk_token": "<unk>"
10
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff