Submit 1x A100 QAT Fix - 1.4078 BPB (Non-Record) [v3]#712
Closed
Shuvam-Banerji-Seal wants to merge 2 commits intoopenai:mainfrom
Closed
Submit 1x A100 QAT Fix - 1.4078 BPB (Non-Record) [v3]#712Shuvam-Banerji-Seal wants to merge 2 commits intoopenai:mainfrom
Shuvam-Banerji-Seal wants to merge 2 commits intoopenai:mainfrom
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
Adds a new non-record submission folder capturing a single-A100 QAT tuning run intended to avoid a torch.quantile compile-time slowdown and to fit a 10-minute wallclock training constraint.
Changes:
- Adds a full
train_gpt.pysnapshot implementing QAT-in-CastedLinear, mixed int6/int8 export, and sliding-window eval. - Adds run artifacts (
train.log) and metadata (submission.json) for the recorded result. - Adds a short README describing the motivation and reported metrics.
Reviewed changes
Copilot reviewed 3 out of 4 changed files in this pull request and generated 1 comment.
| File | Description |
|---|---|
| records/track_non_record_16mb/2026-03-23_Single_A100_QAT_FastFix/train_gpt.py | Training + export script for the submission (QAT, quant/export, eval). |
| records/track_non_record_16mb/2026-03-23_Single_A100_QAT_FastFix/train.log | Captured training/eval log for the run. |
| records/track_non_record_16mb/2026-03-23_Single_A100_QAT_FastFix/submission.json | Leaderboard metadata for the submission. |
| records/track_non_record_16mb/2026-03-23_Single_A100_QAT_FastFix/README.md | Submission overview and claimed results. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
records/track_non_record_16mb/2026-03-23_Single_A100_QAT_FastFix/train_gpt.py
Outdated
Show resolved
Hide resolved
…ix/train_gpt.py Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
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.
Single-device (A100) run tuning hyperparams down from multi-device scales to ensure proper LR scheduling.
torch.quantileforw.amax().clamp_minto evade a 30x compiler performance penalty in Triton.Closes #707