led-bartpho-word-base-16384 / tokenizer_config.json
bluenguyen's picture
Upload 7 files
ce983cc
{
"add_prefix_space": false,
"bos_token": "<s>",
"cls_token": "<s>",
"eos_token": "</s>",
"errors": "replace",
"mask_token": "<mask>",
"model_max_length": 16384,
"name_or_path": "vinai/bartpho-word-base",
"pad_token": "<pad>",
"sep_token": "</s>",
"special_tokens_map_file": null,
"tokenizer_class": "BartTokenizer",
"trim_offsets": true,
"unk_token": "<unk>"
}