span-marker-bert-germeval14 / tokenizer_config.json
stefan-it's picture
model/tokenizer: add initial version
9f149e3
{
"add_prefix_space": true,
"clean_up_tokenization_spaces": true,
"cls_token": "[CLS]",
"do_basic_tokenize": true,
"do_lower_case": false,
"entity_max_length": 8,
"marker_max_length": 128,
"mask_token": "[MASK]",
"max_len": 512,
"model_max_length": 256,
"never_split": null,
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "BertTokenizer",
"unk_token": "[UNK]"
}