finding
active
finding:logit-attribution-heatmaps-and-max-activating-examples-disagree-for-features-425104-and-349410Logit attribution heatmaps and max-activating examples disagree for Features 425104 and 349410
Limitation.
Source paper
extracted_fromNeighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Speculative explanation.
Questions (1)
question
- Unclear reason stated by authors.
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.
- Input-output mapping per feature across ~100 prompts.
- Abstract parity feature.
- Feature 425104 heatmap promotes outputs above ~60 with little input dependence, e.g. promotes 62 given 90finding0.793Diverges from Hanna et al. heatmaps.
- More general greater-than feature.
- Structured number feature.
- Narrow-range greater-than feature.
- Headline result (Figure 1).
- Demonstrates that activation similarity can diverge from logit weight similarity due to interference