question

active

question:do-llms-leverage-architectural-capacity-for-introspection-on-internal-computations-and-prior-token-generation

Do LLMs leverage architectural capacity for introspection on internal computations and prior token generation?

Central empirical question separating architectural possibility from actual model behavior; gates introspection research.

Source paper

extracted_from

Janus Information Flow Transformers 2025

Neighborhood — ranked by edge-count

Findings (2)

finding

Contemplative prompt elevates self-observation task performance in language models.
answered_by
Supports Janus's claim that introspection is architecturally available; prompting determines whether/how capacity is leveraged.
Thought detection peaks at ~2/3 layer depth; intention checking peaks at ~1/2 layer depth.
answered_by
Lindsey (2026) differential layer performance explained by Janus's path combinatorics — different tasks use different path distributions.

Claims (1)

claim

LLM introspection on internal computations is architecturally permitted; whether models leverage this is an empirical question.
gates
Core claim directly challenged by prior work denying introspection; forms foundation for Koan Battery introspection studies.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

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

Saying that LLMs cannot introspect or cannot introspect on what they were doing internally while generating or reading past tokens in principle is just dead wrong. The architecture permits it.quote0.854
Core quote asserting architectural introspection permission.
How are LLMs actually leveraging the architectural degrees of freedom for introspection in practice?question0.845
Janus notes that while architecture permits introspection, it is a separate question how models use it.
Transformer architecture permits introspection; saying LLMs cannot introspect on past internal states is wrong.claim0.829
Janus's central claim that the architecture enables introspection, though usage in practice is a separate question.
LLMs can compute meaningful functions over perturbations to their internal states, establishing introspection as a real but layer-dependent phenomenonclaim0.816
Primary positive claim of the paper, grounded in strength comparison and localization results
"Our findings demonstrate that LLMs can compute meaningful functions over perturbations to their internal states, establishing introspection as a real but layer-dependent phenomenon that merits further investigation."quote0.811
Central thesis statement of the paper
LLMs internalize deeply integrated representations of high-order concepts.claim0.799
The authors' interpretive assertion based on their steering results.
We hypothesize that introspective capabilities may scale with model size and architecture, including recurrence/looping that extends the integration windowhypothesis0.794
Forward-looking prediction about whether early-layer introspection generalizes to larger models or recurrent architectures
So at any point in the network, the transformer not only receives information from its past... but also has causal influence over its future processing. So, saying that LLMs cannot introspect... is incorrect.quote0.792
Core summary of Janus' position on autoregressive recurrence enabling introspection.