Set `clean_up_tokenization_spaces` to false (#8)

- Set `clean_up_tokenization_spaces` to false (85dd5375ce2f6f88c8a81d63b241995d3162c679)
- Set tokenizer class to `PreTrainedTokenizerFast` (df29f94fd88e76ed91c625866c236d725a95de85)


Co-authored-by: Joshua <[email protected]>

tokenizer_config.json

@@ -8005,9 +8005,9 @@
     }
   },
   "bos_token": "<s>",
-  "clean_up_tokenization_spaces": true,
+  "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
   "model_max_length": 1000000000000000019884624838656,
-  "tokenizer_class": "GPT2Tokenizer",
+  "tokenizer_class": "PreTrainedTokenizerFast",
   "unk_token": "<unk>"
 }