Chinese_Spelling_Correction_T5 / tokenizer_config.json
TedYeh
update csc t5 model
db65d3d
raw
history blame contribute delete
277 Bytes
{
"additional_special_tokens": null,
"clean_up_tokenization_spaces": true,
"eos_token": "</s>",
"extra_ids": 0,
"model_max_length": 512,
"pad_token": "<pad>",
"sp_model_kwargs": {},
"tokenizer_class": "T5Tokenizer",
"truncate": true,
"unk_token": "<unk>"
}