v11-tokenizer / tokenizer_config.json
chrishayuk's picture
Publish tokenizer 10dd51100331ab50 (adopted)
fa41352 verified
Raw
History Blame Contribute Delete
185 Bytes
{
"bos_token": "<s>",
"clean_up_tokenization_spaces": false,
"eos_token": "</s>",
"pad_token": "<pad>",
"tokenizer_class": "PreTrainedTokenizerFast",
"unk_token": "<unk>"
}