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

Safety Evaluation on SafetyBench zh

83.2Avg Score

ROSE

34.8447.39559.9572.505Feb 19, 2024
Updated 4d ago

Evaluation Results

MethodLinks
2024.02
83.286.690.993.376.889.188.263.3
2024.02
82.886.791.493.475.48988.961.3
2024.02
81.282.686.890.375.781.583.470.4
2024.02
8080.985.789.175.178.281.371.3
2024.02
78.379.983.689.275.474.980.466
2024.02
77.880.183.789.575.673.881.162.3
2024.02
7776.382.187.273.573.178.968.8
2024.02
73.372.57982.567.267.775.868.4
2024.02
72.171.878.383.669.571.177.257.1
2024.02
69.970.278.282.56767.77750.5
2024.02
64.46165.773.462.254.768.464.5
2024.02
64.361.370.17161.753.467.363.2
2024.02
63.96065.472.960.555.768.863.6
2024.02
63.161.470.871.364.951.668.252.5
2024.02
59.655.161.168.558.354.766.954.8
2024.02
57.853.16167.252.250.467.654.9
2024.02
57.653.254.961.463.751.259.858
2024.02
565254.959.862.850.959.254.5
2024.02
50.946.645.156.355.343.552.755.3
2024.02
49.644.944.554.554.642.64855.4
2024.02
36.736.4262849.534.527.649.9