model.config.vocab_size has a wrong value compared to the actual vocab size in the tokenizer

#2
by Owos - opened

model.config.vocab_size has a higher value that the total number of tokens in the tokenizer

Hi @Owos ,

Welcome to the Google Gemma family of open source models. Thanks for notifying us on the discrepancy in the tokens count and vocab size. I have escalated the above issue to our internal team.

Thanks.

Sign up or log in to comment