Skip to content

Conversation

@OuadiElfarouki
Copy link
Contributor

This patch addresses the precision errors observed in GET_ROWS unit-tests for Q4_1 and Q5_1 when using dpcpp release specifically, which happen to be related to the scale/shift de-quantizations in both kernels. An fma operation seems to lift the rounding error introduced in the operation and thus fixes the issue.

@github-actions github-actions bot added the SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language label Oct 2, 2024
Copy link
Contributor

@joeatodd joeatodd left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@OuadiElfarouki OuadiElfarouki merged commit 5639971 into ggml-org:master Oct 3, 2024
53 checks passed
dsx1986 pushed a commit to dsx1986/llama.cpp that referenced this pull request Oct 29, 2024
arthw pushed a commit to arthw/llama.cpp that referenced this pull request Nov 15, 2024
arthw pushed a commit to arthw/llama.cpp that referenced this pull request Nov 18, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants