Blind human check: rater picks steered image in 100 of 100 pairs in pre-registered SAE re-check
kevrussell · x · 2026-10-27
As part of an independent pre-registered re-check of Goodfire's "painting with concepts" claim, a blind human rater picked the steered image in 100 of 100 pairs against a pass bar written and sealed beforehand. Two deviations from the original plan, including a change of rater, are documented publicly.
Related event: Independent Replication Confirms Goodfire's SAE Feature Steering Claims(3 posts)→
More from Research
- Bret Victor's 2006 essay Magic Ink goes viral again: interactivity is a design crutch — round · 2026-10-27
- ICLR's new Informativeness review question wins praise: "What did I learn?" — CSProfKGD · 2026-10-27
- Citation analysis: neuroscience-AI bridge is real but narrow and tool-heavy — EricTopol · 2026-10-27
- HarborRL open-sources agentic RL: 200+ tasks, 40+ harnesses, one reproducible pipeline — Dear_Treat3688 · 2026-10-27
- Open-source Merlin model from JHU+Stanford auto-segments tumors on CT in one click — MaziyarPanahi · 2026-10-27
- Optimistic take on AI in math: researcher argues the old world is gone — Alex__007 · 2026-10-27