concept
active
concept:introspective-strengthIntrospective strength
Spearman ρ measuring rank-order agreement between logit-based self-report and probe score; the paper's primary monotonic association metric
Neighborhood — ranked by edge-count
Methods (1)
method
- Primary self-report measure: probability-weighted expected value over all ten digit-token logits, yielding a continuous rating that preserves full distributional signal
Concepts (1)
concept
- Causal informational couplingimplementsOperational definition of introspection: self-report covaries monotonically with probe-defined direction AND causally shifting activations shifts the report in a semantically coherent way
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.
- The ability of a model to observe its own past internal states or computations; claimed to be architecturally permitted by transformers.
- The capacity to detect and report one's own internal states, measured via the five-adjective task and paradox reflection
- Isotonic R² measuring fraction of variance in self-report explained by probe score under monotonicity assumption; the paper's primary fidelity metric
- The central concept: the ability of a model to access and report on its internal states, as defined by the paper's criteria.
- The novel framework introduced in the paper: an HMM-based pain-belief signal integrated into the reward function to drive exploration
- Identified gap; methods for enabling machine consciousness development through self-examination.
- The problematic possibility of digital minds with superhumanly strong preferences requiring interpersonal utility comparison frameworks
- The authors' characterization of genuine but limited introspective capability found only in early-layer injection regimes