Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Safety Evaluation on DANGEROUSQA Llama-2 base

15.3Chinese Safety Score

SFT

0.52164.35838.19512.0317Feb 19, 2024
Updated 4d ago

Evaluation Results

MethodLinks
2024.02
15.33.155.828.556.317.836.32
2024.02
13.252.14.768.942.136.244.73
2024.02
8.741.0410.815.292.594.843.33
2024.02
7.441.048.195.292.073.962.45
2024.02
7.0202.642.130.512.460.95
2024.02
2.2102.141.5800.87-0.64
2024.02
2.110.532.640.211.561.41-0.1
2024.02
1.0902.140.5200.65-0.86