JosephusCheung commited on
Commit
a0fec25
1 Parent(s): 915124b

Upload folder using huggingface_hub

Browse files
Files changed (4) hide show
  1. merges.txt +0 -0
  2. special_tokens_map.json +5 -0
  3. tokenizer_config.json +8 -10
  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 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "unk_token": "<|endoftext|>"
5
+ }
tokenizer_config.json CHANGED
@@ -1,11 +1,9 @@
1
  {
2
- "remove_space": false,
3
- "do_lower_case": false,
4
- "tokenizer_class": "QWenTokenizer",
5
- "auto_map": {
6
- "AutoTokenizer": [
7
- "tokenization_qwen.QWenTokenizer",
8
- null
9
- ]
10
- }
11
- }
 
1
  {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<|endoftext|>",
4
+ "tokenizer_class": "GPT2Tokenizer",
5
+ "clean_up_tokenization_spaces": true,
6
+ "eos_token": "<|endoftext|>",
7
+ "model_max_length": 1000000000000000019884624838656,
8
+ "unk_token": "<|endoftext|>"
9
+ }
 
 
vocab.json ADDED
The diff for this file is too large to render. See raw diff