vilt_finetuned_200 / tokenizer_config.json
aqachun's picture
Training in progress, step 200
38dce15
raw
history blame
351 Bytes
{
"clean_up_tokenization_spaces": true,
"cls_token": "[CLS]",
"do_lower_case": true,
"mask_token": "[MASK]",
"model_max_length": 40,
"pad_token": "[PAD]",
"processor_class": "ViltProcessor",
"sep_token": "[SEP]",
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "BertTokenizer",
"unk_token": "[UNK]"
}