LLMs perform in-context learning as trajectories through a structured low-dimensional conceptual belief space, with the structure visible in both behavior and internal representations and causally manipulable via interventions.
What Features in Prompts Jailbreak LLMs? Investigating the Mechanisms Behind Attacks , shorttitle =
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 4roles
background 1polarities
background 1representative citing papers
Causal mediation analysis shows harmful LLM outputs arise in late layers from MLP failures and gating neurons, with early layers handling harm context detection and signal propagation.
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.
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.
citing papers explorer
-
Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space
LLMs perform in-context learning as trajectories through a structured low-dimensional conceptual belief space, with the structure visible in both behavior and internal representations and causally manipulable via interventions.
-
Why Do Large Language Models Generate Harmful Content?
Causal mediation analysis shows harmful LLM outputs arise in late layers from MLP failures and gating neurons, with early layers handling harm context detection and signal propagation.
-
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.