urdu_sms_hingbert / tokenizer_config.json
tanaya-b's picture
Hingbert model finetuned on WASSA 2023 data
f1e528f
{
"cls_token": "[CLS]",
"do_basic_tokenize": true,
"do_lower_case": false,
"mask_token": "[MASK]",
"max_len": 512,
"name_or_path": "l3cube-pune/hing-bert",
"never_split": null,
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"special_tokens_map_file": "/opt/ravindra.n/misc/models/BERT/special_tokens_map.json",
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "BertTokenizer",
"unk_token": "[UNK]"
}