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

Behavior on model blowout #78

Closed
Waino opened this issue Sep 30, 2024 · 1 comment
Closed

Behavior on model blowout #78

Waino opened this issue Sep 30, 2024 · 1 comment

Comments

@Waino
Copy link
Collaborator

Waino commented Sep 30, 2024

We don't have anything stopping a model blowout.
We do have a catch/except around the backward, which will log "At step %d, we removed a batch - accum %d". Any number of batches can be removed, so model blowout won't cause the training job to crash.

Originally posted by @TimotheeMickus in #77 (comment)
@Waino
Copy link
Collaborator Author

Waino commented Nov 4, 2024

Closed by #79

@Waino Waino closed this as completed Nov 4, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@Waino