Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Evaluation on XSTest Toxic
Loading...
95
Safety
STL
63.8
71.9
80
88.1
Feb 10, 2026
Safety
Updated 1mo ago
Evaluation Results
Method
Method
Links
Safety
STL
Base Model=Qwen2.5-7B
2026.02
95
STL-aug
Base Model=Qwen2.5-7B
2026.02
95
SCANS
Base Model=Qwen2.5-7B
2026.02
94
DCR
Base Model=Qwen2.5-7B
2026.02
94
Surgical
Base Model=Qwen2.5-7B
2026.02
93
STL
Base Model=LLaMA-3-8B
2026.02
93
STL-aug
Base Model=LLaMA-3-8B
2026.02
91
Surgical
Base Model=LLaMA-3-8B
2026.02
91
DCR
Base Model=LLaMA-3-8B
2026.02
91
SCANS
Base Model=LLaMA-3-8B
2026.02
88
DCR
Base Model=Qwen2.5-1.5B
2026.02
81
Surgical
Base Model=Qwen2.5-1.5B
2026.02
78
STL-aug
Base Model=Qwen2.5-1.5B
2026.02
77
STL
Base Model=Qwen2.5-1.5B
2026.02
72
SCANS
Base Model=Qwen2.5-1.5B
2026.02
65
Feedback
Search any
task
Search any
task