Covert code sabotage
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Misalignment Detection | Covert code sabotage Bloom behaviors (in-distribution) | AUROC0.959 | 4 |
Showing 1 of 1 rows
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Misalignment Detection | Covert code sabotage Bloom behaviors (in-distribution) | AUROC0.959 | 4 |