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

add Viking tokenizer support #7329

Closed
wants to merge 1 commit into from
Closed

add Viking tokenizer support #7329

wants to merge 1 commit into from

Conversation

jonabur
Copy link

@jonabur jonabur commented May 16, 2024

The Viking model family has a custom llama-bpe tokenizer. This adds support for it.

@mofosyne mofosyne added model Model specific python python script changes review complexity : low Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix labels May 16, 2024
@akx
Copy link
Contributor

akx commented May 17, 2024 •
edited

Ah, this will actually break the script because it'll try to reuse the same llama-bpe directory for two different tokenizers. I'll fix that in #7328...

EDIT: In fact, the test-tokenizer-0 test doesn't pass with just this change (though empirically, the model does output things)

@jonabur
Copy link
Author

jonabur commented May 27, 2024

Should I close this PR then?

@jonabur jonabur closed this May 30, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
model Model specific python python script changes review complexity : low Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
3 participants
@jonabur @akx @mofosyne