arabic-relation-extraction / tokenizer_config.json
edchengg
First model version for testing pipeline
0b65a98
raw
history blame
562 Bytes
{
"cls_token": "[CLS]",
"do_basic_tokenize": true,
"do_lower_case": true,
"mask_token": "[MASK]",
"name_or_path": "lanwuwei/GigaBERT-v4-Arabic-and-English",
"never_split": null,
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"special_tokens_map_file": "/srv/share5/ychen3411/huggingface_cache/c8a919731ad7551df88a744a8399190ea80a9961eaf96b8a59f86ef617e5144f.6a644b330fad284f98393c5832f71bce43df6855fa6ac7c9e44ed6271b708170",
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "BertTokenizer",
"unk_token": "[UNK]"
}