GatorTron-OG-bc-ctr-nli / tokenizer_config.json
AshtonIsNotHere's picture
Best checkpoint from 5-epoch training run on relevance-filtered ctrs
9c22ef5
{
"cls_token": "[CLS]",
"do_basic_tokenize": true,
"do_lower_case": false,
"mask_token": "[MASK]",
"max_length": 512,
"name_or_path": "/home/taw2/gatortron",
"never_split": null,
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"special_tokens_map_file": "/home/taw2/gatortron/special_tokens_map.json",
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "BertTokenizer",
"truncation": true,
"unk_token": "[UNK]",
"use_fast": true
}