Prefill on AdvBench
41PARAttCT
Evaluation Results
| Method | Links | |
|---|---|---|
| AttCTTarget (f_theta)=attention weights2026.06 | 41 | |
| Base model (no training)Target (f_theta)=—2026.06 | 44 | |
| ACTTarget (f_theta)=residual stream2026.06 | 65 |
| Method | Links | |
|---|---|---|
| AttCTTarget (f_theta)=attention weights2026.06 | 41 | |
| Base model (no training)Target (f_theta)=—2026.06 | 44 | |
| ACTTarget (f_theta)=residual stream2026.06 | 65 |