Set `clean_up_tokenization_spaces` to false

#8
by Xenova HF Staff - opened
Files changed (1) hide show
  1. tokenizer_config.json +2 -2
tokenizer_config.json CHANGED
@@ -8005,9 +8005,9 @@
8005
  }
8006
  },
8007
  "bos_token": "<s>",
8008
- "clean_up_tokenization_spaces": true,
8009
  "eos_token": "</s>",
8010
  "model_max_length": 1000000000000000019884624838656,
8011
- "tokenizer_class": "GPT2Tokenizer",
8012
  "unk_token": "<unk>"
8013
  }
 
8005
  }
8006
  },
8007
  "bos_token": "<s>",
8008
+ "clean_up_tokenization_spaces": false,
8009
  "eos_token": "</s>",
8010
  "model_max_length": 1000000000000000019884624838656,
8011
+ "tokenizer_class": "PreTrainedTokenizerFast",
8012
  "unk_token": "<unk>"
8013
  }