tempclin-biobertpt-clin / tokenizer_config.json
lisaterumi's picture
Upload tokenizer_config.json
745ca19
raw
history blame
No virus
622 Bytes
{
"cls_token": "[CLS]",
"do_basic_tokenize": true,
"do_lower_case": false,
"mask_token": "[MASK]",
"max_len": 512,
"max_length": 256,
"name_or_path": "pucpr/biobertpt-clin",
"never_split": null,
"pad_token": "[PAD]",
"padding": true,
"sep_token": "[SEP]",
"special_tokens_map_file": "/root/.cache/huggingface/transformers/6ba8c26d43da6320fddc3c2d9e164fe97894eaa8fd10359985c4a74d74031b86.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d",
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "BertTokenizer",
"truncation": true,
"unk_token": "[UNK]"
}