Skip to content

for intel xpu case, use MatMul8bitFp even not use ipex #1728

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
Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

for intel xpu case, use MatMul8bitFp even not use ipex #1728

wants to merge 1 commit into from

Conversation

kaixuanliu
Copy link

For Intel XPU case, use MatMul8bitFp is faster than MatMul8bitLt. And it can avoid the datatype overflow issue in L105

@yao-matrix
Copy link

@matthewdouglas , could u pls help review, it will make int8 LoRA finetuning loss overflow to nan on XPU, thx very much.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@kaixuanliu @yao-matrix