fix(e5): move the mtp head to cuda with the student - #117
Merged
Conversation
build_mtp ran while the student was still on cpu (student.to('cuda')
happens after label generation), so the head's Linears stayed on cpu
and the first training step died in a device mismatch. Labels survive
in the run dir — the relaunch skips labeling and trains.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
build_mtp ran while the student was still on cpu, so the head's Linears stayed there and the first training step crashed with a device mismatch (labels are safe in the run dir; relaunch skips labeling). Second launch-iteration bug on E5 — the run is resume-safe by construction.