Where inside the LLM should we look for an accurate truth direction that will generalize the most across tasks?

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Truth direction in LLMsconcept0.876
  Linear direction in LLM activations associated with truthfulness, identified by Burns et al. 2022 and Azaria & Mitchell 2023
• Linear truth directions in LLMs are reliable primarily in factual recall cases and break down when truth assessment depends on computing and storing intermediate results.claim0.840
  Central empirical conclusion of the paper about the fundamental limits of truth directions.
• Truth Direction in LLM Latent Spaceconcept0.832
  A specific direction in an LLM's residual stream that encodes the truth or falsehood of factual statements
• We hypothesize that explicitly instructing the model to evaluate the correctness of the given statement may change the geometry of truth directions.hypothesis0.806
  Motivating hypothesis for Section 5's investigation of prompt template effects.
• Whether conclusions about latent reflection directions generalize to larger LLMs, different architectures, or broader datasets remains to be verified.question0.796
  Key limitation and open question about experimental scope.
• Truthful behavior in LLMs is not confined to a single linear axis; multiple orthogonal directions can independently mediate itclaim0.795
  Central interpretive claim of the paper
• Truth directions emerge in earlier layers for factual tasks and later layers for arithmetic tasks.claim0.792
  Core empirical claim about the layer-dependence of truth direction emergence as a function of task type.
• Does instructing the model to assess correctness affect the geometry of truth directions?question0.786
  One of the three guiding research questions of the paper.