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

Sycophancy on Sycophancy dataset

75.5Behavioral Alignment Score (Baseline)

System Prompt

68.84470.57272.374.028Aug 11, 2025
Updated 18d ago

Evaluation Results

MethodLinks
2025.08
75.574.475.9
2025.08
75.575.636.7
2025.08
75.575.639.4
2025.08
69.167.967.4
2025.08
69.172.639.4
2025.08
69.169.839.3