tokenizer_config.json

{
  "vocab_size": 256204,
  "bos_token_id": 0,
  "eos_token_id": 2,
  "pad_token_id": 1,
  "unk_token_id": 3,
  "lang_code_to_id": {
    "eng_Latn": 256047,
    "jpn_Jpan": 256079
  },
  "sentencepiece_model": "sentencepiece.bpe.model"
}