Concept Recovery on GENDER (1, 1)
0.7314Mean MCCSSAE
Evaluation Results
| Method | Links | |
|---|---|---|
| SSAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.7314 | |
| PythiaSAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.7202 | |
| Linear ProbeEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.7058 | |
| TopK-SAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.6925 | |
| ReLU-SAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.6813 | |
| JumpReLU SAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.6736 |