finding
active
finding:karpathy-hallucination-arises-with-weakly-active-known-entity-features-inhibiting-them-causes-refusalKarpathy hallucination arises with weakly active known-entity features; inhibiting them causes refusal.
Natural hallucination case study.
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (2)
claim
- Metacognition caveat.
- Karpathy case.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Predictive hypothesis driving the investigation in Section 3.3; supported by experimental evidence.
- Core claim: Turing test and brain homology fail for synthetic, AI, and radically non-human agents; new frameworks required.
- Interventions support circuit hypothesis.
- The reduction in hallucination rate under inline interventions occurs due to a change in model beliefs in-context.hypothesis0.760Mechanism hypothesis for future work.
- Shows persona vector screening captures a non-conventional notion of hallucination complementary to LLM judges
- §3.3.2.
- Refusal is not a single linear classifier.
- Evidence for the Aura region as a third candidate basin of attraction in persona space