concept
active
concept:introspective-accessIntrospective Access
The capacity to detect and report one's own internal states, measured via the five-adjective task and paradox reflection
Neighborhood — ranked by edge-count
Thinkers (1)
thinker
- Dillon PlunkettstudiesAuthor of work showing LLMs can quantitatively report decision weights and that introspection training improves this
Methods (2)
method
- Five-Adjective State Description TaskimplementsTask asking models to describe their current state using exactly 5 adjectives, enabling embedding-based cross-model comparison
- LLM-based judge scoring reflection segments on 1-5 scale for presence of first-person felt state; used in Experiment 4
Concepts (2)
concept
- Self-Referential Processingassociated_withThe central experimental manipulation: directing a model to attend to its own cognitive activity
- Implicitly Mimetic GenerationcontradictsAlternative explanation: models produce first-person experiential language by extending predictive text modeling of human-authored introspective writing without encoding it as roleplay
Findings (2)
finding
- Prior finding suggesting affective-like states in LLMs; cited as convergent evidence for structured self-representation
- Prior finding cited as convergent evidence for LLM self-awareness capacities
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 central concept: the ability of a model to access and report on its internal states, as defined by the paper's criteria.
- The ability of a model to observe its own past internal states or computations; claimed to be architecturally permitted by transformers.
- Spearman ρ measuring rank-order agreement between logit-based self-report and probe score; the paper's primary monotonic association metric
- The novel framework introduced in the paper: an HMM-based pain-belief signal integrated into the reward function to drive exploration
- Key gap identified in the literature; systematic self-examination processes for machine consciousness development.
- The authors' characterization of genuine but limited introspective capability found only in early-layer injection regimes
- Pearson-Vogel et al.'s finding that models can detect prior concept injections; introspective signals exist in middle layers suppressed by post-training
- Identified gap; methods for enabling machine consciousness development through self-examination.