finding
active
finding:predictive-component-s-v-sp-similarity-is-invariant-across-garden-path-and-control-0-47-vs-0-44-while-saes-and-novel-code-vary-strongly-e-g-topk-0-09-vs-0-38Predictive component's V→SP similarity is invariant across garden path and control (0.47 vs 0.44), while SAEs and novel code vary strongly (e.g., TopK 0.09 vs 0.38).
Fig. 8b,c.
Source paper
extracted_from(2025) · Lubana, Ekdeep Singh · Rager, Can · Hindupur, Sai Sumedh R. · Costa, Valerie +12
Neighborhood — ranked by edge-count
Papers (1)
paper
Hypotheses (1)
hypothesis
- If sentence ambiguity is reflected in latent codes, V–SP similarity should be similar for garden path and control sentences while V–OP similarity is higher.associated_withsupportsPrediction for garden-path test.
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.
- Fig. 7c over 50 GPT-5 stories.
- Fig. 26.
- Fig. 8a.
- Fig. 7d,e.
- Table 3.
- Table 3.
- Mechanistic evidence for two distinct representational pathways
- Appendix E replication of DIM alignment finding in Qwen model