Concept Recovery on SYCOPHANCY
0.9827Mean MCCLinear Probe
Evaluation Results
| Method | Links | |
|---|---|---|
| Linear ProbeEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.9827 | |
| PythiaSAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.4931 | |
| TopK-SAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.4735 | |
| JumpReLU SAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.4511 | |
| ReLU-SAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.4347 | |
| SSAEEvaluation Protocol=Recovered decoder directions evaluated on paired observations (f(x), f(x˜))2025.02 | 0.4303 |