DAS behavioral loss achieves IIA of 0.997±0.001 on synthetic 10-class dataset training/test sets

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• DAS behavioral loss produces EMD along feature dimensions of 0.032±0.003 on synthetic 10-class datasetfinding0.854
  Quantitative baseline for divergence using behavioral DAS loss on synthetic dataset
• Modified CL loss achieves IIA of 0.9988±0.0005 on synthetic 10-class dataset training/test setsfinding0.832
  IIA for modified CL loss on synthetic dataset, comparable to behavioral DAS
• For small CL loss weights epsilon, IIA is maintained (potentially improved) while EMD decreases in Boundless DAS on a 7B LLMfinding0.784
  Empirical result showing the CL loss can reduce divergence without sacrificing interpretability accuracy
• Modified CL loss outperforms behavioral DAS loss in OOD transfer from dense to sparse class partitionfinding0.783
  Key practical utility result: CL loss improves generalization of alignment to out-of-distribution settings
• DAS achieves 100% IIA on hierarchical equality task with |N|=16, intervention size 8, Layer 1finding0.767
  DAS discovers a perfect alignment between the feed-forward network and the Both Equality Relations high-level model.
• DAS on randomly initialized small networks (|N|=16) achieves only 0.50 IIA (chance), cannot construct new behaviorsfinding0.758
  Demonstrates DAS cannot manufacture behaviors from random structure in appropriately sized networks.
• DAS learning rate of 5e-3 outperforms 1e-3 (used in Wu et al. 2023) for small training sets in CausalGymfinding0.750
  Hyperparameter tuning result for DAS; different from prior work due to smaller training set size
• SAE training loss (MSE + L1 penalty with decoder norm scaling)method0.743
  The objective function combining L2 reconstruction error and L1 penalty scaled by decoder norm, used to train the SAE.