LLM Activations
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Downstream Utility Evaluation | LLM Activations | Sparse Probing Accuracy87.9 | 8 | |
| Feature Interpretability | LLM Activations | AutoInterp Score86.9 | 8 | |
| Hierarchical Feature Alignment | LLM Activations | Absorption98.8 | 8 | |
| Sparse Reconstruction | LLM Activations | L049.4 | 8 |
Showing 4 of 4 rows