scaling01 says its 'dirty loopers' posts pushed labs to report the behavior in system cards
scaling01 · x · 2026-11-10
- The account scaling01 says Neel Nanda's earlier post on NCRI/NCKI drew its attention, after which it began posting about "dirty loopers" — looping/idling behavior patterns in models on long-horizon tasks.
- It claims that every lab's system cards now report the behavior, viewing this as concrete evidence that community discussion has improved model evaluation transparency.
Related event: KOL Claims Credit for Getting "Dirty Loopers" Into Model System Cards(2 posts)→
More from Models
- Claude Fable 5.2 tops VIALS-AA science visual reasoning benchmark at 67% with heavy tool use — ArtificialAnlys · 2026-11-10
- Dev builds a mobile game with Fable 5.2 as a Claude Artifact in under an hour — ChrisGPT · 2026-11-10
- SpaceX says Grok detects aerospace foreign object debris with 99.9% accuracy vs humans' 94% — elonmusk · 2026-11-10
- Fable 5.2's first ASCII backrooms art is so good people suspect it's programmatic — repligate · 2026-11-10
- Community pushes AI labs to publish system cards as machine-readable jsonl data — scaling01 · 2026-11-10
- ARLoop: fully-open 1.2B looped transformer beats Huginn-3.5B and Qwen3.5-2B on all 8 benchmarks — hhsun1 · 2026-11-10