Off-policy training data for LLM behavior probes causes significant generalization failures especially for intent-based behaviors like deception, and performance on coerced incentivised data correlates with real on-policy success.
Propositional interpretability in artificial intelligence
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
Persona axes derived from contrastive prompts and PCA yield linear probes that generalize better than raw-activation probes across 10 datasets for deception and sycophancy.
The paper claims that mechanistic interpretability needs philosophy as a partner to clarify concepts, refine methods, and navigate epistemic and ethical complexities in AI systems.
citing papers explorer
-
The Impact of Off-Policy Training Data on Probe Generalisation
Off-policy training data for LLM behavior probes causes significant generalization failures especially for intent-based behaviors like deception, and performance on coerced incentivised data correlates with real on-policy success.
-
Do Linear Probes Generalize Better in Persona Coordinates?
Persona axes derived from contrastive prompts and PCA yield linear probes that generalize better than raw-activation probes across 10 datasets for deception and sycophancy.
-
Mechanistic Interpretability Needs Philosophy
The paper claims that mechanistic interpretability needs philosophy as a partner to clarify concepts, refine methods, and navigate epistemic and ethical complexities in AI systems.