mdeberta-v3-base-ctebmsp / tokenizer_config.json
GuillemGSubies's picture
Upload the tokenizer
6e9811e
raw
history blame
529 Bytes
{
"bos_token": "[CLS]",
"cls_token": "[CLS]",
"do_lower_case": false,
"eos_token": "[SEP]",
"keep_accents": true,
"mask_token": "[MASK]",
"model_max_len": 512,
"model_max_length": 1000000000000000019884624838656,
"name_or_path": "microsoft/mdeberta-v3-base-32-4e-05-0.2-ctebmsp_best",
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"sp_model_kwargs": {},
"special_tokens_map_file": null,
"split_by_punct": false,
"tokenizer_class": "DebertaV2Tokenizer",
"unk_token": "[UNK]",
"vocab_type": "spm"
}