finding
active
finding:in-the-co-activation-toy-model-features-with-low-co-activation-probabilities-appear-exclusive-and-symmetric-across-modelsIn the co-activation toy model, features with low co-activation probabilities appear exclusive and symmetric across models
Supports the co-activation explanation of symmetry.
Source paper
extracted_fromSiddharth Mishra-Sharma · Trenton Bricken · Jack Lindsey · Adam Jermyn +4
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Plausible explanation.
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.
- Variant toy model where latents from a common pool have co-activation probabilities, yielding symmetric exclusive features.
- Symmetry is not explained by the base toy model.
- Fig. 4 caption.
- Validates toy model.
- Low-cosine shared features are best interpreted as same features used differently in the two modelsclaim0.771Interpretation of cosine distribution.
- Superposition controls robustness.
- Key claim enabling the virtual instance view to survive server changes
- Figure 4 left.