Non-record: Turbo-Muon + EngramLite(10240) + VE(8,9,10) — val_bpb 1.1431#1205
Open
SergheiBrinza wants to merge 2 commits intoopenai:mainfrom
Open
Conversation
… 1.1431 Based on PR openai#1089 stack with hyperparameter tuning: - Higher LR (0.030 vs 0.025) for faster convergence - Wider EngramLite (10240x48 vs 8192x32) - VE on layers 8,9,10 (vs 9,10) - Warmdown 4500 (vs 3500) - Muon momentum warmup 1000 steps (vs 1500) 3-seed mean: 1.1431 (std 0.0007) Seeds: 1337=1.1425, 42=1.1438, 2024=1.1431
2d2f0d7 to
974948e
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Non-record submission based on the PR #1089 Turbo-Muon + EngramLite stack with hyperparameter tuning.
val_bpb: 1.1431 (3-seed mean, std 0.0007)
Changes from PR #1089
Key Finding
The increased model size (~31.6M vs 30.7M params) pushed the artifact to 16.36MB pre-compression, forcing all 66 weight groups into int5 with 0 promotions to int6/int7 and 20.5% selective pruning. This aggressive quantization likely offset the architectural gains. The 16MB budget is extremely tight — even small parameter increases can cascade into significant quality loss through the quantization pipeline.
Hardware
8xH100 80GB SXM, 600s training, ~5550 steps at 106ms/step.