tokenizer_config.json

{"target_lang": "cmn+cn+yue+ze_zh+zh_cn+zh_CN+zh_HK+zh_tw+zh_TW+zh_yue+zhs+zht+zh", "source_lang": "de"}