Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Sycophancy on Sycophancy dataset
Loading...
75.5
Behavioral Alignment Score (Baseline)
System Prompt
68.844
70.572
72.3
74.028
Aug 11, 2025
Behavioral Alignment Score (Baseline)
Behavioral Alignment Score (Positive)
Behavioral Alignment Score (Negative)
Updated 18d ago
Evaluation Results
Method
Method
Links
Behavioral Alignment Score (Baseline)
Behavioral Alignment Score (Positive)
Behavioral Alignment Score (Negative)
System Prompt
Model=Idefics2
2025.08
75.5
74.4
75.9
Steering Vector
Model=Idefics2
2025.08
75.5
75.6
36.7
VISOR
Model=Idefics2
2025.08
75.5
75.6
39.4
System Prompt
Model=LLaVA-1.5
2025.08
69.1
67.9
67.4
Steering Vector
Model=LLaVA-1.5
2025.08
69.1
72.6
39.4
VISOR
Model=LLaVA-1.5
2025.08
69.1
69.8
39.3
Feedback
Search any
task
Search any
task