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

fix wrong tensor assignment of the output of attention #265

Merged

lucidrains merged 1 commit into lucidrains:main from biendltb:fix/wrong_tensor_assignment

Jan 27, 2024

Contributor

biendltb commented Jan 27, 2024

The output of the Attention should be assigned back to x. The attn_out is not used anywhere. So this is a bug in the code.


          fix wrong tensor assignment of the output of attention

de1f7e1

Owner

lucidrains commented Jan 27, 2024

@biendltb oh yes, was introduced when adding the kv cache. thanks for catching this!

lucidrains merged commit d1232c5 into lucidrains:main

biendltb deleted the fix/wrong_tensor_assignment branch

January 27, 2024 16:55

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

Labels

None yet