HeackMT5-ZhCleanText1ML / tokenizer_config.json
heack's picture
initial release
cfb349e
raw
history blame contribute delete
311 Bytes
{
"additional_special_tokens": null,
"clean_up_tokenization_spaces": true,
"eos_token": "</s>",
"extra_ids": 0,
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<pad>",
"sp_model_kwargs": {},
"tokenizer_class": "T5Tokenizer",
"tokenizer_file": null,
"unk_token": "<unk>"
}