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

Fixed QLoRA with mutli-gpu and reduce CUDA memory pressure during eval #3486

Merged
merged 2 commits into from
Jul 29, 2023
Merged

Fixed QLoRA with mutli-gpu and reduce CUDA memory pressure during eval #3486

merged 2 commits into from
Jul 29, 2023

Conversation

tgaddair
Copy link
Collaborator

No description provided.

arnavgarg1
arnavgarg1 approved these changes Jul 29, 2023
View reviewed changes
Copy link
Contributor

@arnavgarg1 arnavgarg1 left a comment

Choose a reason for hiding this comment

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

LGTM since we're only doing device_map=auto with quantisation

@tgaddair tgaddair mentioned this pull request Jul 29, 2023
Closed
@tgaddair tgaddair changed the title Fix qlora with mutli gpu Fixed QLoRA with mutli-gpu and reduce CUDA memory pressure during eval Jul 29, 2023
@github-actions
Copy link

Unit Test Results

  6 files  ±0    6 suites  ±0   1h 6m 32s ⏱️ + 3m 21s
34 tests ±0  29 ✔️ ±0    5 💤 ±0  0 ❌ ±0 
88 runs  ±0  72 ✔️ ±0  16 💤 ±0  0 ❌ ±0 

Results for commit 3bea9be. ± Comparison against base commit 62a5524.

@tgaddair tgaddair merged commit 053f17c into master Jul 29, 2023
14 of 15 checks passed
@tgaddair tgaddair deleted the multi-gpu-llm branch July 29, 2023 18:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@arnavgarg1 arnavgarg1 arnavgarg1 approved these changes

@geoffreyangus geoffreyangus Awaiting requested review from geoffreyangus geoffreyangus is a code owner

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
@tgaddair @arnavgarg1