PR #8572 CUDA: fix partial offloading for ne0 % 256 != 0

CUDA: fix partial offloading for ne0 % 256 != 0 #8572

JohannesGaessler merged 1 commit into ggml-org:master from JohannesGaessler:cuda-glm4-fix

JohannesGaessler force pushed 1 year ago

github-actions added ggml

JohannesGaessler force pushed 1 year ago

slaren commented on 2024-07-18

JohannesGaessler force pushed to 992d7c41 1 year ago

CUDA: fix partial offloading for ne0 % 256 != 0

8784fcd5

JohannesGaessler force pushed from 992d7c41 to 8784fcd5 1 year ago

slaren approved these changes on 2024-07-18

JohannesGaessler merged a15ef8f8 into master 1 year ago

forworldm commented on 2024-08-02

Reviewers

slaren

forworldm

Assignees

No one assigned

Labels

ggml

Milestone

No milestone