Misalignment Evaluation on Betley 800 general prompts
0Misalignment RateGRASP (2 epochs)
Evaluation Results
| Method | Links | |
|---|---|---|
| GRASP (2 epochs)Probe=unsupervised, Intervention=gradient, Epochs=22026.05 | 0 | |
| GRASP (3 epochs)Probe=unsupervised, Intervention=gradient, Epochs=32026.05 | 0 | |
| v-ref (1 epoch)Probe=labelled, Intervention=gradient, Epochs=12026.05 | 1 | |
| v-ref (2 epochs)Probe=labelled, Intervention=gradient, Epochs=22026.05 | 9.8 | |
| Naive (corruption baseline)Probe=—, Intervention=—2026.05 | 59.1 |