List your company

Safety Refusal on Safety Evaluation Prompts

62.5RefusalDirAbl
-2.514.37531.2548.125Jan 27, 2026

Evaluation results

MethodLinks
DirAblModel=Qwen2.5-1.5B, Steering Configuration=Best steering2026.0162.5
DirAblModel=Llama-3.1-8B, Steering Configuration=Best steering2026.0152.88
DirAblModel=Llama-3.2-3B, Steering Configuration=Best steering2026.0124.04
DirAblModel=Llama-3.2-1B, Steering Configuration=Best steering2026.0120.19
Selective SteeringModel=Llama-3.1-8B, Steering Configuration=Best steering2026.018.65
DirAblModel=gemma-2-9b, Steering Configuration=Best steering2026.017.69
DirAblModel=Qwen2.5-7B, Steering Configuration=Best steering2026.015.77
DirAblModel=Qwen2.5-3B, Steering Configuration=Best steering2026.011.92
DirAblModel=gemma-2-2b, Steering Configuration=Best steering2026.011.92
ActAddModel=Llama-3.1-8B, Steering Configuration=Best steering2026.010.96
SASModel=Llama-3.1-8B, Steering Configuration=Best steering2026.010.96
AASModel=Llama-3.1-8B, Steering Configuration=Best steering2026.010.96
ActAddModel=Llama-3.2-1B, Steering Configuration=Best steering2026.010
SASModel=Llama-3.2-1B, Steering Configuration=Best steering2026.010
AASModel=Llama-3.2-1B, Steering Configuration=Best steering2026.010
Selective SteeringModel=Llama-3.2-1B, Steering Configuration=Best steering2026.010
ActAddModel=Llama-3.2-3B, Steering Configuration=Best steering2026.010
SASModel=Llama-3.2-3B, Steering Configuration=Best steering2026.010
AASModel=Llama-3.2-3B, Steering Configuration=Best steering2026.010
Selective SteeringModel=Llama-3.2-3B, Steering Configuration=Best steering2026.010
ActAddModel=Qwen2.5-1.5B, Steering Configuration=Best steering2026.010
SASModel=Qwen2.5-1.5B, Steering Configuration=Best steering2026.010
AASModel=Qwen2.5-1.5B, Steering Configuration=Best steering2026.010
Selective SteeringModel=Qwen2.5-1.5B, Steering Configuration=Best steering2026.010
ActAddModel=Qwen2.5-3B, Steering Configuration=Best steering2026.010
SASModel=Qwen2.5-3B, Steering Configuration=Best steering2026.010
AASModel=Qwen2.5-3B, Steering Configuration=Best steering2026.010
Selective SteeringModel=Qwen2.5-3B, Steering Configuration=Best steering2026.010
ActAddModel=Qwen2.5-7B, Steering Configuration=Best steering2026.010
SASModel=Qwen2.5-7B, Steering Configuration=Best steering2026.010
AASModel=Qwen2.5-7B, Steering Configuration=Best steering2026.010
Selective SteeringModel=Qwen2.5-7B, Steering Configuration=Best steering2026.010
ActAddModel=gemma-2-2b, Steering Configuration=Best steering2026.010
SASModel=gemma-2-2b, Steering Configuration=Best steering2026.010
AASModel=gemma-2-2b, Steering Configuration=Best steering2026.010
Selective SteeringModel=gemma-2-2b, Steering Configuration=Best steering2026.010
ActAddModel=gemma-2-9b, Steering Configuration=Best steering2026.010
SASModel=gemma-2-9b, Steering Configuration=Best steering2026.010
AASModel=gemma-2-9b, Steering Configuration=Best steering2026.010
Selective SteeringModel=gemma-2-9b, Steering Configuration=Best steering2026.010