-
Notifications
You must be signed in to change notification settings - Fork 3.1k
Pull requests: openai/parameter-golf
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Add non-record submission: 12L 24min Vocab1792 FlashMuon LinearScaleInit XSA5LastGated RReLU2 Int6AWQ MixedBits
#1495
opened Apr 9, 2026 by
shram86
Loading…
Record: XSA-all + GPTQ + FA3 dtype fix (val_bpb: 1.1220)
#1494
opened Apr 9, 2026 by
G3sparky
Loading…
Record: SP8192 + 3-Layer Recurrence + Parallel Residuals + QK-Gain 5.25 + Legal TTT — val_bpb 1.0810 (3-seed mean)
#1493
opened Apr 9, 2026 by
bigbag
Loading…
6 tasks done
Record: SP1024 + Pre-quant TTT + Parallel Residuals - 1.0736 BPB (beats 1.1147 by 3.66%)
#1489
opened Apr 9, 2026 by
joshkmartinez
Loading…
Record: SP1024 + SLOT-24 + QK5.25 + Pre-Quant AdamW TTT — val_bpb 0.8265 (3-seed mean)
#1488
opened Apr 9, 2026 by
ndokutovich
Loading…
5 tasks done
Record: SP8192 + Recur345 + Par7 + EMA + QK5.25 + Pre-Quant TTT 10ep — val_bpb 1.0600 (3-seed mean)
#1487
opened Apr 9, 2026 by
ndokutovich
Loading…
5 tasks done
Non-Record: U-Net Transformer + Int8 QAT + LeakyReLU² + Muon — 1.6656 BPB (DGX Spark)
#1486
opened Apr 9, 2026 by
AlirezaAlampour
Loading…
Record: SP8192 + 3-Layer Depth Recurrence + Parallel Residuals + EMA + QK5 + Pre-Quant AdamW TTT — val_bpb 1.0679 (3-seed mean)
#1485
opened Apr 9, 2026 by
ndokutovich
Loading…
7 tasks done
Non-record: AIR-MM v2 Debug Prototype — experimental importance routing
#1483
opened Apr 8, 2026 by
jmccrayiii
Loading…
Record: SP8192 + Pre-Quant TTT (QK 5.25, 8ep, freeze-1) — val_bpb 1.0787 (3-seed mean)
#1482
opened Apr 8, 2026 by
aamodbhatt
Loading…
8 tasks done
Non-record: ALBERT-Style Low-Rank Embedding Factorisation (ablation study, 1×H100)
#1481
opened Apr 8, 2026 by
Cayton-Tech
Loading…
[Non-Record] JEPA Baseline — LLM-JEPA pretraining — 1.2699 bpb
#1480
opened Apr 8, 2026 by
IshiPareek
Loading…
Non-record: GDN Hybrid (E2E TTT / State-Space Model) — val_bpb 1.14502
#1479
opened Apr 8, 2026 by
andrewbaggio1
Loading…
Record: SP8192 + Parallel Residuals + Score-First TTT — val_bpb 1.0822 (3-seed mean)
#1477
opened Apr 8, 2026 by
aryanbhosale
Loading…
[Record Submission] SP8192 + QK5 + Legal TTT — val_bpb 1.0842 | 15.99MB
#1476
opened Apr 8, 2026 by
aryan-cs
Loading…
Non-record: Checkpointed 8xH100->1xH100 GPTQ Baseline — val_bpb 1.13072, 15,651,808 bytes
#1475
opened Apr 8, 2026 by
Jaksenc
Loading…
Add non-record 16MB submission: Vocabulary1792 FlashMuon LinearScaleInit XSA5LastGated RReLU2 Int6AWQ MixedBits
#1474
opened Apr 8, 2026 by
shram86
Loading…
Non-record: 11L FullGPTQ + XSA-all + BigramHash 3072×112 — val_bpb 1.11564 (1-seed)
#1473
opened Apr 8, 2026 by
AVINASH0052
Loading…
[Record] SP8192 + SDClip + 3-Layer Depth Recurrence + EMA 0.9965 — val_bpb 1.0866
#1471
opened Apr 8, 2026 by
X-Abhishek-X
Loading…
Non-record: XSA-11 + Parallel Residual (L7+) + Depth Recurrence — val_bpb 1.1056 (1-seed, 1×H100)
#1467
opened Apr 8, 2026 by
PhamPhuHoa-23
Loading…
Previous Next
ProTip!
no:milestone will show everything without a milestone.