[Non-Record] Extended Compute Scaling Analysis: 1.0853 BPB at 50K steps (11.5 hours) on 4×A100MIG#1005
Open
OnlyJundong wants to merge 1 commit intoopenai:mainfrom
Open
Conversation
…50K steps (11.5 hours) on 4xA100 MIG
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
This submission is a non-record submission. It studies how the current record-track SOTA (PR #549 by @abaybektursun) scales under extended compute, removing the 10-minute wall-clock constraint. The same architecture and code are trained for 20K–50K steps (5.5-11.5 hours training) on 4×A100 MIG instances (approximately 10× slower per step than 8×H100 SXM).
Results
Best run: 50K steps and 11.5 hours (4×A100 MIG, seed 1337)
20K steps and 5.5 hours (4×A100 MIG, 2-seed comparison)
Plots
BPB vs Steps (ASCII plot)
Artifact Size vs Steps (ASCII plot)