tokenizer_config.json

{
  "added_tokens_decoder": {},
  "backend": "sentencepiece",
  "bos_token": "<s>",
  "clean_up_tokenization_spaces": true,
  "cls_token": "<s>",
  "eos_token": "</s>",
  "extra_special_tokens": [
    "__java__",
    "__python__",
    "__en_XX__"
  ],
  "is_local": false,
  "language_codes": "base",
  "local_files_only": false,
  "mask_token": "<mask>",
  "model_max_length": 200,
  "pad_token": "<pad>",
  "sep_token": "</s>",
  "sp_model_kwargs": {},
  "src_lang": null,
  "tgt_lang": null,
  "tokenizer_class": "PLBartTokenizer",
  "unk_token": "<unk>"
}