RuModernBERT-Intent-Classifier / tokenizer_config.json
galkinv42's picture
Upload tokenizer
455cbeb verified
raw
history blame contribute delete
555 Bytes
{
"backend": "tokenizers",
"clean_up_tokenization_spaces": true,
"cls_token": "[CLS]",
"extra_special_tokens": [
"<|padding|>",
"<|endoftext|>",
"[UNK]",
"[CLS]",
"[SEP]",
"[PAD]",
"[MASK]"
],
"is_local": false,
"mask_token": "[MASK]",
"model_input_names": [
"input_ids",
"attention_mask"
],
"model_max_length": 1000000000000000019884624838656,
"model_specific_special_tokens": {},
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"tokenizer_class": "TokenizersBackend",
"unk_token": "[UNK]"
}