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.

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

Add MFU Monitoring to LLM #56

Merged
merged 32 commits into from
Dec 30, 2022
Merged

Conversation

vchiley
Copy link
Contributor

@vchiley vchiley commented Dec 21, 2022

Adds MFU to SpeedMonitor
A generalized variant of this feature should eventually be integrated into Composer's SpeedMonitor callback, but this will work well for MosaicGPT in examples/llm.

Screenshot 2022-12-21 at 1 54 21 PM

Screenshot 2022-12-21 at 1 55 49 PM

llm/main.py Outdated Show resolved Hide resolved
Copy link
Contributor

@abhi-mosaic abhi-mosaic left a comment

Choose a reason for hiding this comment

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

Approving to unblock, but I think it's important that we don't just hardcode the 312e12 number, otherwise when people run this repo on alternative hardware they will get incorrect MFU values in their reporting.

llm/src/speed_monitor_w_mfu.py Outdated Show resolved Hide resolved
llm/src/speed_monitor_w_mfu.py Outdated Show resolved Hide resolved
Copy link
Contributor

@abhi-mosaic abhi-mosaic left a comment

Choose a reason for hiding this comment

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

LGTM! Excited to have all this info auto generated.

@abhi-mosaic abhi-mosaic merged commit fd08771 into mosaicml:main Dec 30, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants