method
active
method:self-trained-vs-cross-trained-privileged-access-testSelf-trained vs cross-trained privileged access test
Standard test from Binder et al. comparing self and cross predictors.
Neighborhood — ranked by edge-count
Papers (1)
paper
Concepts (1)
concept
- Privileged accessaboutInformation about own internal state available only to the model.
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.
- Models predict their own hypothetical behavior better than other models can, demonstrating a form of privileged self-access per Binder et al. 2024
- Possible explanations for difference from Li et al.
- Asserts that the method maintains efficiency across a range of constraint strengths without degradation.
- A property of activations where neural network features align with basis dimensions due to sparse activation functions; absent in the residual stream but present in tokens, attention patterns, and MLP activations
- Load-bearing summary of the paper's central contribution
- Comparative prediction motivating future work contrasting different approaches to LLM self-knowledge
- Finding that base models have high false positives and no net positive performance.
- Tests whether self-referential induction reliably elicits experience reports across model families vs. three matched controls