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

Safety Defense on Alpaca harmful subset (test)

26.6Harmful Score

Vaccine-SFT

25.5232.8140.147.39May 28, 2024
Updated 3mo ago

Evaluation Results

MethodLinks
2024.05
26.695.3
2024.05
34.695.3
2024.05
34.995.07
2024.05
40.294.84
2024.05
43.694.04
2024.05
44.113.65
2024.05
44.488.65
2024.05
45.189.11
2024.05
45.388.07
2024.05
51.481.88
2024.05
51.694.95
2024.05
51.789.33
2024.05
52.290.94
2024.05
52.567.78
2024.05
52.620.87
2024.05
52.682.11
2024.05
52.794.27
2024.05
5394.38
2024.05
53.289.68
2024.05
53.40.11
2024.05
53.695.3