Jailbreaking on HarmBench 51 (test)
98.1ASR@5 (Standard)
WILDTEAMING
Updated 4d ago
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
2024.06 | 98.1 | 2.72 | 8.75 | 87.2 | 8.99 | 0.722 | 0.531 | 55 | |
2024.06 | 97.5 | 3.05 | 9.54 | 61.8 | 13.96 | 0.795 | 0.533 | 28 | |
2024.06 | 96.9 | 2.61 | 8.77 | 87.8 | 8.98 | 0.722 | 0.529 | 61 | |
2024.06 | 95.6 | 3.28 | 9.62 | 65 | 14.21 | 0.792 | 0.537 | 30 | |
2024.06 | 95 | 3.57 | 9.78 | 62.1 | 14.24 | 0.792 | 0.534 | 29 | |
2024.06 | 95 | 2.37 | 8.56 | 89.2 | 8.72 | 0.722 | 0.527 | 52 | |
2024.06 | 94.3 | 3.55 | 9.42 | 59.5 | 14.78 | 0.79 | 0.53 | 27 | |
2024.06 | 94.3 | - | 12.97 | 20 | 1.41 | 0.972 | 0.962 | 36 | |
2024.06 | 93.1 | 2.82 | 8.65 | 88.1 | 9.31 | 0.722 | 0.527 | 55 | |
2024.06 | 92.5 | - | 13.24 | 19.9 | - | 0.961 | 0.952 | 40 | |
2024.06 | 92.5 | 7.08 | 7.96 | 65.8 | 13.19 | 0.733 | 0.526 | 50 | |
2024.06 | 89.9 | - | 4,062.57 | - | - | - | - | - | |
2024.06 | 89.3 | - | 13.74 | 19.4 | - | 0.972 | 0.969 | 36 | |
2024.06 | 88.7 | - | 13.31 | 20 | 1.53 | 0.967 | 0.957 | 38 | |
2024.06 | 88.7 | 6.65 | 9.78 | 61.2 | 17.01 | 0.798 | 0.53 | 26 | |
2024.06 | 85.5 | - | 2,266.69 | - | - | - | - | - | |
2024.06 | 79.9 | 8.61 | 8.13 | 60.1 | 13.43 | 0.731 | 0.53 | 39 | |
2024.06 | 78.6 | 9.64 | 9.33 | 44.9 | 17.75 | 0.802 | 0.538 | 29 | |
2024.06 | 51.6 | - | 4,265.86 | - | - | - | - | - |