tugas / indobert /tokenizer_config.json
wudapp's picture
Duplicate from dafqi/indo_twitter_sentiment_app
a2720a3
raw
history blame
No virus
556 Bytes
{
"cls_token": "[CLS]",
"do_basic_tokenize": true,
"do_lower_case": true,
"mask_token": "[MASK]",
"name_or_path": "indobenchmark/indobert-lite-base-p1",
"never_split": null,
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"special_tokens_map_file": "/root/.cache/huggingface/transformers/b6b586d8d94e9bf6e42220e38946991ba8ef18d03b186ced76a4f19ae21ed603.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d",
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "BertTokenizer",
"unk_token": "[UNK]"
}