In 17 LLMs from five families, the apparent scaling of concept steerability vanishes under normalized, held-out controls, while a decodable world map persists and other neuroscience parallels are measurement-dependent.
Science , volume=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
When Is a Steerable Concept Representation Real? Measurement Confounds in a Cross-Family Audit of Neuroscience Parallels in LLMs
In 17 LLMs from five families, the apparent scaling of concept steerability vanishes under normalized, held-out controls, while a decodable world map persists and other neuroscience parallels are measurement-dependent.