-
Notifications
You must be signed in to change notification settings - Fork 2.7k
Pull requests: openai/parameter-golf
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Non-record: 24.7M params · int6 · Binary U-Net/SmearGate/BigramHash · 1.5hr · RTX 5060 Ti 16GB
#997
opened Mar 28, 2026 by
randy06122001-boop
Loading…
Pre-Enrichment + EMA-GPU + SmearGate + XSA4 (val_bpb=1.1478, …
#996
opened Mar 28, 2026 by
Idan3011
Loading…
Record: 1.0362 BPB — SGD Momentum 0.95 TTT + HedgeMixer + Per-Layer LR
#995
opened Mar 28, 2026 by
dexhunter
Loading…
Add Kshitij submission (1x H100, val_bpb 1.4315, env-based config)
#994
opened Mar 28, 2026 by
singhaikshitijjain
Loading…
Record: 33.6M Int5 GPTQ + Score-First TTT (val_bpb=1.1145, 3-seed)
#991
opened Mar 28, 2026 by
ibarrajo
Loading…
5 tasks done
ClownCar: Frugendorff compression baseline + canonical DeltaNet integration
#990
opened Mar 28, 2026 by
newjordan
Loading…
QAT x SWA Ablation: SWA sabotages QAT (-3.64 mBPB, 3-seed validated)
#989
opened Mar 27, 2026 by
alexanderaperry-arch
Loading…
Record: Packed N-gram + Two-Pass Dirichlet CTW — val_bpb 0.0830 (3-seed mean)
#986
opened Mar 27, 2026 by
sofiabod
Loading…
9 tasks done
Add 128-cluster baseline submission files
#985
opened Mar 27, 2026 by
danielweidinger2299-debug
Loading…
submission 2026-03-27_PhaseCoherenceGatedGradients PIC-GID + ParallelMuon
#984
opened Mar 27, 2026 by
jzgdev
Loading…
Non-record: 11L LeakyReLU(0.5)^2 + EMA + Int6 Quantization
#983
opened Mar 27, 2026 by
WHITELOTUS0
Loading…
4 tasks
Non-record: Sliding Patch Attentions + MoE (2-layer compact run)
#981
opened Mar 27, 2026 by
BurguerJohn
Loading…
[LOGOS-44] Holographic Coherence Architecture - Loss 0.9377 - 8.88 MB
#980
opened Mar 27, 2026 by
slowomir33-arch
Loading…
Record: 1.1387 BPB — 11L LeakyReLU² + Early [email protected] + GPTQ-lite + EMA
#979
opened Mar 27, 2026 by
0xadvait
Loading…
Review: Rerun of #972 with actual full-vocab normalization
#978
opened Mar 27, 2026 by
AnirudhRahul
Loading…
2 tasks done
LeakyReLU(0.75)² + Legal TTT + Parallel Muon — 1.1185 BPB (3-seed mean)
#977
opened Mar 27, 2026 by
michaelwinczuk
Loading…
Add 1.20 BPB submission with Legal TTT and Calibration (9L/448D)
#976
opened Mar 27, 2026 by
Vibes-me
Loading…
Non-record: QNA + SQWA compression thesis (8xH100 SXM)
#975
opened Mar 27, 2026 by
Abhishek8108
Loading…
Non-record: Random Linear Map Adapter Projections — 1.21MB artifact (val_bpb=1.6542)
#974
opened Mar 27, 2026 by
anthony-maio
Loading…
Non-record: BESE Novel Tokenizer — 38-Token Structured Alphabet + BPE, 288 Vocab, 12.9MB
#973
opened Mar 27, 2026 by
mrbese
Loading…
Non-record: GatedDeltaNet SSM via fla library — 1.2907 bpb, 15.79MB
#970
opened Mar 27, 2026 by
dnldsz
Loading…
Record: Order-20 Dirichlet Posterior + Phrase Cache — 0.11545 BPB (3-seed)
#968
opened Mar 27, 2026 by
dentity007
Loading…
5 tasks done
Record: 1.0450 BPB — SGD TTT + HedgeMixer with Per-Layer LR Groups
#967
opened Mar 27, 2026 by
dexhunter
Loading…
Previous Next
ProTip!
Adding no:label will show everything without a label.