Record: Scylla + Full GPTQ + XSA-all + FA3 — val_bpb 0.9485 (3-seed mean)#1184
Merged
cocohearts merged 1 commit intoopenai:mainfrom Apr 23, 2026
Merged
Record: Scylla + Full GPTQ + XSA-all + FA3 — val_bpb 0.9485 (3-seed mean)#1184cocohearts merged 1 commit intoopenai:mainfrom
cocohearts merged 1 commit intoopenai:mainfrom
Conversation
sunnypatneedi
pushed a commit
to sunnypatneedi/parameter-golf
that referenced
this pull request
Mar 31, 2026
- logs/daily_research.md: append 2026-03-31 research section - PR openai#771 CLOSED (score-first TTT rule violation) - PR openai#727 CLOSED (n-gram illegal — no renormalization) - Merged SOTA: 1.1147 (PR openai#1019, 2026-03-25) - New PRs: openai#1184 (0.9485 Scylla tokenizer), openai#1185 (0.9641) - SLOT eval technique, Full GPTQ, QK-Gain 4.0 documented - CLAUDE.md: update Competition Strategy + lessons 21-24 - Merged SOTA updated to 1.1147 - Current Best Path rewritten for 2026-03-31 - Lessons openai#21-24: TTT fix, n-gram risk, Scylla, SLOT - TTT constraint clarified to score-first protocol - Version bumped to v9.0 https://claude.ai/code/session_015z6QKyKzDSYzTniW1GPhAe
Contributor
Author
|
Byte accounting uses the identical candidate.meta.npz from PR #1143. I have stronger results under SLOT and with additional sweeps. Unsure if SLOT will be removed under rule 3 so I’ll hold off on additional submissions. |
5 tasks
andrewbaggio1
added a commit
to andrewbaggio1/parameter-golf
that referenced
this pull request
Apr 2, 2026
…asurement error PR openai#1184's 0.9485 BPB becomes 1.1289 with corrected byte accounting. 93% of the gap is byte denominator inflation, not model quality. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Contributor
Author
|
Closing due to byte accounting bug in candidate.meta.npz, will wait for direction on SLOT and rerun before re-submitting. |
This was referenced Apr 4, 2026
sunnypatneedi
pushed a commit
to sunnypatneedi/parameter-golf
that referenced
this pull request
Apr 24, 2026
…ai#1787 Polar Express NS new base; PR openai#1795 PPM 1.01252; Issue openai#1604 deadline passed; Session 20 - Merged SOTA 1.0810 confirmed Day 15 (README not updated despite Scylla record commit) - Scylla 0.9485 committed to track_10min_16mb/ on Apr 23 (PR openai#1184) but byte accounting disputed by PR openai#1271 (corrected ~1.1289 bpb); treat merged SOTA as 1.0810 - PR openai#771 CLOSED/REJECTED confirmed; PR openai#727 CLOSED (illegal); PR openai#758 open but dead; PR openai#731 still awaiting seeds 1337+2024 - Issue openai#1604 (CaseOps ruling): NO @valerio-oai response in 11 days; self-deadline Apr 24 passed; proceed with clean legal stack immediately - NEW: PR openai#1787 (nprime06, 1.06335) — new community-consensus clean base with Polar Express Newton-Schulz (arXiv:2505.16932, ICLR 2026) + MIN_LR=0.10 warmdown floor - NEW: PR openai#1795 (OE-GOD, 1.01252) — byte-level PPM order-4 adaptive mixture; gate legality concern fixed; await organizer ruling before implementing - NEW: PR openai#1797 (dexhunter, 1.06157) — PR openai#1787 + SmearGate + LQER Asym; new dexhunter best - NEW: PR openai#1802 (aamodbhatt, 1.0771) — Polar Express NS + Multi-Phase Global TTT - TECHNIQUE: Polar Express NS (arXiv:2505.16932) and Gram NS (Dao-AILab) added to table - TECHNIQUE: MIN_LR=0.10 warmdown floor added to best-stack approach - Updated competition strategy: stop waiting for CaseOps, implement clean stack with Polar Express NS + MIN_LR immediately (6 days to deadline) https://claude.ai/code/session_01JZ3FiS937NwLHt3Fv9WHPD
sunnypatneedi
pushed a commit
to sunnypatneedi/parameter-golf
that referenced
this pull request
Apr 26, 2026
…1835 PPM-D 1.00136 new watch; NgramRes stackable; Day 17 plateau; Session 22 - Upstream commit 7427de2 (Alex Zhao, OpenAI Apr 26): Scylla 0.9485 (PR openai#1184) removed as invalid record; PR openai#1813 (djeidy Scylla 0.94166) effectively dead by proxy - PR openai#1835 (anmarhindi, 1.00136): PPM-D order-5 byte mixture, binary-λ gate, score-first, 15,993,020 bytes — most credible extraordinary claim yet; wait 24h for community BPB check - PR openai#1834 (ghrua, 1.08034): NgramRes 3-gram MLP +0.6M params + sliding-window attn layers 0-3 — modest, stackable - PR openai#731 (Hedge Mixer): still OPEN, 2 seeds pending, no merge - Merged SOTA 1.0810 definitively confirmed; target ≤1.0760; 4 days to deadline https://claude.ai/code/session_01XbdTRT7zPHoGp3LfQV4yXF
hilbertmeng
pushed a commit
to hilbertmeng/parameter-golf
that referenced
this pull request
Apr 30, 2026
…ean) (openai#1184) Co-authored-by: icryo <icryo@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
2 participants
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
3-Seed Results
Key Innovation
Scylla tokenizer (998 tokens, @simon-marcus PR #1143) + modern training stack:
PR #1143 used the old SOTA base. This submission applies the modern stack to the same tokenizer, yielding 12% better BPB.
Test plan
Credits