mt5_extend_py2500 / tokenizer_config.json
Kajiura Teruno
Expanded vocabulary by adding 2,500 words
4c1c18a
{
"additional_special_tokens": [],
"clean_up_tokenization_spaces": true,
"eos_token": "</s>",
"extra_ids": 0,
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<pad>",
"sp_model_kwargs": {},
"tokenizer_class": "T5Tokenizer",
"unk_token": "<unk>"
}