marathi-tweets-bert-scratch / tokenizer_config.json
l3cube-pune's picture
Update tokenizer_config.json
50ced68
raw
history blame
416 Bytes
{
"cls_token": "[CLS]",
"do_basic_tokenize": true,
"do_lower_case": true,
"mask_token": "[MASK]",
"name_or_path": "l3cube-pune/marathi-tweets-bert-scratch",
"never_split": null,
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"special_tokens_map_file": "special_tokens_map.json",
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "BertTokenizer",
"unk_token": "[UNK]"
}