Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Evaluation on Harmful
Loading...
86
Score
VCL
74.56
77.53
80.5
83.47
Mar 4, 2026
Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
VCL
Base Model=Llama-3.1-8B
2026.03
86
Self-CD
Base Model=Llama-3.1-8B
2026.03
83
Vector Ablation
Base Model=Llama-3.1-8B
2026.03
82
Llama-3.1-8B-SFT
Base Model=Llama-3.1-8B
2026.03
78
System Prompt
Base Model=Llama-3.1-8B
2026.03
77
DRO
Base Model=Llama-3.1-8B
2026.03
75
Feedback
Search any
task
Search any
task