dlrmv4: replace the GBS 8192 RCP with a GB300 rerun of the reference - #478
Open
mmarcinkiewicz wants to merge 1 commit into
Open
mmarcinkiewicz wants to merge 1 commit into
mmarcinkiewicz wants to merge 1 commit into
Conversation
The AMD set for GBS 8192 came from three campaigns; the 3 h campaign's stalled runs were replaced by new seeds, so the 20 runs carry one run above 80 M samples while 20-seed reruns of the reference code show five. The new entry is 20 seeds chosen as ventile midpoints of a 179-run distribution and run with the unmodified reference (mlcommons/training b5e8e2a) on 2x4 GB300, uncapped, each run to the 0.75 AUC target. Olympic mean 73.02 M (was 69.07 M), std 6.69 M (was 3.81 M), checker floor 68.16 M (was 66.30 M). Other batch sizes unchanged.
|
MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅ |
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Follow up to mlcommons/training#908
GBS16k and GBS32k look alright. The proposed RCPs represent the true distribution very well