tacomet-gpt2-xl-japanese / tokenizer_config.json
Eiki's picture
Upload tokenizer
9fb9e82 verified
raw
history blame contribute delete
245 Bytes
{
"bos_token": "<s>",
"clean_up_tokenization_spaces": true,
"eos_token": "</s>",
"model_max_length": 1024,
"padding_side": "right",
"tokenizer_class": "PreTrainedTokenizerFast",
"truncation_side": "right",
"unk_token": "<unk>"
}