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

[Bug] Llama 3 DynamicNTK RoPE is incorrect #152

Closed

awni opened this issue Nov 2, 2024 · 2 comments

Assignees

Member

awni commented Nov 2, 2024

It's basically just wrong.. but works well enough for short sequences that it's not noticeable. We should follow the Python implementation.

davidkoski self-assigned this

Contributor

DePasqualeOrg commented Nov 15, 2024

Sorry about that. I think that was my mistake. I've fixed this in #154.

Collaborator

davidkoski commented Dec 10, 2024

This is merged

davidkoski closed this as completed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment