{ "backend": "tokenizers", "bos_token": "", "eos_token": "", "jieba": true, "model_max_length": 512, "pad_token": "", "pinyin_format": "pinyin-code", "tokenizer_class": "EncodedMandarinTokenizerFast", "transliteration": "pinyin-code", "unk_token": "", "use_jieba": true, "auto_map": { "AutoTokenizer": [ "tokenization_pinyin_code.EncodedMandarinTokenizer", "tokenization_pinyin_code.EncodedMandarinTokenizerFast" ] }, "tokenizer_kind": "sentencepiece" }