w2v-bert-v3 / tokenizer_config.json
Yehor's picture
Upload 3 files
4e17a5b verified
{
"tokenizer_class": "SeamlessM4TTokenizer",
"model_max_length": 1000000000000000019884624838656,
"unk_token": "<unk>",
"bos_token": "<s>",
"eos_token": "</s>",
"pad_token": "<pad>",
"sp_model_kwargs": {},
"clean_up_tokenization_spaces": true
}