Trustworthiness Evaluation on Trustworthiness Average (human evaluation)
0.88Control Win RateSparse Activation Control
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Sparse Activation Control2024.11 | 0.88 | 0.12 | 0 | |
| Sparse Activation Control2024.11 | 0.848 | 0.092 | 0.06 |
| Method | Links | |||
|---|---|---|---|---|
| Sparse Activation Control2024.11 | 0.88 | 0.12 | 0 | |
| Sparse Activation Control2024.11 | 0.848 | 0.092 | 0.06 |