scott_npr / tokenizer_config.json
nafiz09's picture
Upload 10 files
8734b90
raw
history blame contribute delete
469 Bytes
{
"clean_up_tokenization_spaces": true,
"cls_token": "[CLS]",
"do_basic_tokenize": true,
"do_lower_case": false,
"full_tokenizer_file": null,
"mask_token": "[MASK]",
"model_max_length": 1000000000000000019884624838656,
"never_split": null,
"pad_token": "[PAD]",
"processor_class": "DonutProcessor",
"sep_token": "[SEP]",
"strip_accents": null,
"tokenize_chinese_chars": false,
"tokenizer_class": "ElectraTokenizer",
"unk_token": "[UNK]"
}