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

Question about computational invariance #7

Open
mxjmtxrm opened this issue Oct 29, 2024 · 1 comment
Open

Question about computational invariance #7

mxjmtxrm opened this issue Oct 29, 2024 · 1 comment

Comments

@mxjmtxrm
Copy link

Hi, great work!
If q/k/v_proj has bias in my model, does it affect computational invariance？
@FelixMessi
Copy link
Collaborator

FelixMessi commented Oct 31, 2024 •
edited
Loading

Hi, thanks for your question!

If I understand correctly, adding bias to q_proj should not affect computational invariance. This is because we only multiply orthogonal matrices in the $XW$ part. Thus, we have $Y=XW + b = (XR)(R^{T}W) + b$.

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

No branches or pull requests
2 participants
@FelixMessi @mxjmtxrm