question
active
question:what-constitutes-a-good-explanation-of-model-behaviorWhat constitutes a good explanation of model behavior?
Motivating question.
Source paper
extracted_from(2026) · Adam Karvonen · Euan Ong · Subhash Kantamneni · Samuel Marks
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Central evaluative stance.
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.
- Practical implication.
- Author's interpretation of the VTAB alignment results echoing Tolstoy
- Central challenge for active inference stated in Discussion.
- Tested in D.1.
- Context length ablation.
- Critique of using formal specifications alone for concept definition.
- Key limitation identified: NLAs hallucinate specific details while preserving thematic accuracy; informs practical usage.
- Key consequence: GPT's power comes from simulating something contingent.