Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

(Re)activate test-tokenizer-0-falcon #3233

Closed
goerch opened this issue Sep 17, 2023 · 2 comments · Fixed by #3252
Closed

(Re)activate test-tokenizer-0-falcon #3233

goerch opened this issue Sep 17, 2023 · 2 comments · Fixed by #3252

Comments

@goerch
Copy link
Collaborator

goerch commented Sep 17, 2023

I noticed test-tokenizer-0-falcon seems to be deactivated. Any reasons for this?
@ggerganov
Copy link
Owner

It's failing - no unicode support. So decided to disable it until we add it

@goerch
Copy link
Collaborator Author

goerch commented Sep 17, 2023 •
edited
Loading

I'm trying to get something like test-tokenizer-1 for falcon working. Yeah, Unicode seems to be part of the problem.

@ all concerned : if we want to do proper tokenization, please don't modify the original vocabulary and add

    GGML_ASSERT(vocab.id_to_token.size() == vocab.token_to_id.size());

to your vocab-loading code, otherwise I don't see a chance.

@goerch goerch mentioned this issue Sep 18, 2023
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Work on the BPE tokenizer goerch/llama.cpp
2 participants
@ggerganov @goerch