w-en2zh-otm / tokenizer_config.json
Chun's picture
add tokenizer
7e4aa32
raw
history blame
No virus
567 Bytes
{"src_lang": "en_XX", "tgt_lang": null, "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": "<s>", "language_codes": "ML50", "special_tokens_map_file": "/home/suraj/projects/mbart-50/mbart-50/special_tokens_map.json", "name_or_path": "facebook/mbart-large-50-one-to-many-mmt", "model_max_length": 1024, "sp_model_kwargs": {}, "tokenizer_class": "MBart50Tokenizer"}