wangchanberta-base-att-spm-uncased / tokenizer_config.json
lalital's picture
Change space token from <th_roberta_space_token> to <_>
04ff256
raw
history blame
341 Bytes
{
"pretrained_model_name_or_path": "CamembertTokenizer",
"bos_token": "<s>",
"eos_token": "</s>",
"unk_token": "<unk>",
"sep_token": "</s>",
"cls_token": "<s>",
"pad_token": "<pad>",
"mask_token": "<mask>",
"additional_special_tokens": [
"<s>NOTUSED",
"</s>NOTUSED",
"<_>"
]
}