Non-record: 11L GEPA + 12k Steps + Pure Int6 + Legal TTT (val_bpb=1.1079)#612
Open
Christopher-Lee-McClendon wants to merge 1 commit intoopenai:mainfrom
Conversation
Non-record submission: 11L GEPA architecture trained for 12000 steps (7k peak-LR + 5k warmdown) on 4xA100-40GB with pure int6 per-row quantization using 15-candidate GPTQ-lite clip search and zstd-22 compression. Legal score-first TTT (SGD, 10 epochs, momentum 0.9) drives final BPB from 1.154 (int6 quant) to 1.1079. - Pre-TTT float base: 1.1268 (step 12000) - Post-quant pre-TTT: ~1.154 - Final with legal TTT: 1.10788263 - Artifact: 15.51 MB (14.72 MB model + 78 KB code) - 27M parameters, pure int6 quantization
This was referenced Mar 24, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Non-Record Submission: 11L GEPA + 12k Steps + Pure Int6 + Legal TTT
val_bpb = 1.1079 (1.10788263 exact) | Pre-TTT float: 1.1268 | Int6 quant: ~1.154 | TTT gain: −0.046 | Artifact: 14.79 MB
Summary
GEPA architecture (11L, 27M params) trained for 12,000 steps (7k peak-LR + 5k warmdown) on 4×A100-40GB. Pure int6 per-row quantization with 15-candidate GPTQ-lite clip search + zstd-22 compression. Legal score-first TTT (SGD, momentum 0.9, lr=0.002, 10 epochs, freeze first 2 blocks).
Key Results
Novel Contributions
Track
track_non_record_16mb— unlimited compute, 16 MB artifact limit.Checklist