koen_bbpe64k / tokenizer_config.json
kimsan0622's picture
Upload tokenizer
a584a43
raw
history blame contribute delete
339 Bytes
{
"name_or_path": "vocab/ko_en_44000/ko20000_ByteLevelBPETokenizer",
"special_tokens": [
"<s>",
"</s>",
"<unk>",
"<sep>",
"<pad>",
"<cls>",
"<mask>"
],
"special_tokens_map_file": "vocab/ko_en_44000/ko20000_ByteLevelBPETokenizer/special_tokens_map.json",
"tokenizer_class": "PreTrainedTokenizerFast"
}