method
active
method:binary-detection-taskBinary Detection Task
Task paradigm from prior work asking 'Did you detect an injected thought?' via YES/NO logit comparison; shown here to be confounded
Neighborhood — ranked by edge-count
Papers (1)
paper
Concepts (1)
concept
- global logit shiftassociated_withThe methodological confound identified by this paper: injection biases model toward 'YES' for any binary question regardless of content
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.
- LLM judge classification accuracy measuring interpretability; threshold 0.8.
- Task where a random word is prefilled as the assistant's response, then the model is asked whether it intended to say that word, testing introspection on prior intentions.
- Task of detecting a model's internal thoughts; found by Lindsey (2026) to peak at ~2/3 depth in transformers.
- Fundamental structure (G, M, R) modeling objects with attributes; gives rise to polar maps and concept lattices.
- Flag actions taken for abnormal reasons.
- An LLM-based classifier that returns 1 if response contains a clear subjective experience report and 0 otherwise
- Current research focus in literature; contrasted with the need for systematic introspective processes.
- Heuristic algorithm using detrending, Hann windowing, and FFT to classify token-level limiting behavior as FixedPoint, Orbit, Slider, or Unknown