tokenizer_config.json

{"do_lower_case": false, "subword_tokenizer_type": "wordpiece", "word_tokenizer_type": "mecab", "model_max_length": 512}