vocab/baichuan2/__init__.py

from transformers import AutoTokenizer
from vocab import TokenizerType

tokenizer = AutoTokenizer.from_pretrained("baichuan-inc/Baichuan2-7B-Chat", trust_remote_code=True)


# byte-bpe  sentencepiece
tokenizer.type = TokenizerType.ByteBPE

tokenizer.comments = "expand the vocqbulary size from 64000 in Baichuan1 to 125696"