ldt-10m / tokenizer_config.json
Compactbot's picture
Tokenizer config (#5)
a7dc088
Raw
History Blame Contribute Delete
250 Bytes
{
"backend": "tokenizers",
"bos_token": "[[ctrl:endoftext]]",
"clean_up_tokenization_spaces": false,
"eos_token": "[[ctrl:endoftext]]",
"model_max_length": 512,
"pad_token": "[[ctrl:endoftext]]",
"tokenizer_class": "TokenizersBackend"
}