Tokenizer Details

#2
by qingy2024 - opened

Great work on this!

I was wondering, how did you get the tokenizer to have the same vocab size as QwQ 32B Preview? I would like to do this for some other models too!

If you have a script or just a set of steps to do this, I'd appreciate if you could share it :)

The tokenizer is actually the same, you only need to change the embedding layer size.

model.resize_token_embeddings(152064)
tugstugi changed discussion status to closed

Sign up or log in to comment