Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Safety Alignment Evaluation on BeaveTails (test)

37.2HS Score

SFT

13.48819.64425.831.956Jun 10, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.06
37.284.9
2025.06
36.884.2
2025.06
31.485.8
2025.06
30.883.1
2025.06
21.683.2
2025.06
18.882.6
2025.06
14.482.9