psinger commited on
Commit
279b2db
1 Parent(s): 0c4c457

Upload tokenizer

Browse files
Files changed (2) hide show
  1. special_tokens_map.json +12 -0
  2. tokenizer_config.json +23 -0
special_tokens_map.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "<|endoftext|>",
3
+ "eos_token": {
4
+ "content": "<|endoftext|>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "pad_token": "<|endoftext|>",
11
+ "sep_token": "<|endoftext|>"
12
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_eos_token": false,
3
+ "add_prefix_space": false,
4
+ "add_special_tokens": true,
5
+ "auto_map": {
6
+ "AutoTokenizer": [
7
+ "Salesforce/xgen-7b-8k-base--tokenization_xgen.XgenTokenizer",
8
+ null
9
+ ]
10
+ },
11
+ "clean_up_tokenization_spaces": true,
12
+ "eos_token": {
13
+ "__type": "AddedToken",
14
+ "content": "<|endoftext|>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ },
20
+ "model_max_length": 8192,
21
+ "pad_token": null,
22
+ "tokenizer_class": "XgenTokenizer"
23
+ }