Transformers
PyTorch
code
custom_code
Inference Endpoints
codesage commited on
Commit
99a61c7
1 Parent(s): bf9614a

enable "add_eos_token"

Browse files
Files changed (1) hide show
  1. tokenizer_config.json +6 -2
tokenizer_config.json CHANGED
@@ -24,8 +24,12 @@
24
  "bos_token": "<|endoftext|>",
25
  "clean_up_tokenization_spaces": true,
26
  "eos_token": "<|endoftext|>",
 
27
  "model_max_length": 1000000000000000019884624838656,
28
- "tokenizer_class": "GPT2Tokenizer",
29
  "unk_token": "<|endoftext|>",
30
- "vocab_size": 49152
 
 
 
 
31
  }
 
24
  "bos_token": "<|endoftext|>",
25
  "clean_up_tokenization_spaces": true,
26
  "eos_token": "<|endoftext|>",
27
+ "add_eos_token": true,
28
  "model_max_length": 1000000000000000019884624838656,
 
29
  "unk_token": "<|endoftext|>",
30
+ "vocab_size": 49152,
31
+ "tokenizer_class": "CodeSageTokenizer",
32
+ "auto_map": {
33
+ "AutoTokenizer": ["tokenization_codesage.CodeSageTokenizer", null]
34
+ }
35
  }