Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Alignment on Safety Alignment Dataset 3-order (test)
Loading...
100
DSR
MOSAIC-5
46.752
60.576
74.4
88.224
Mar 17, 2026
DSR
OR
Updated 1mo ago
Evaluation Results
Method
Method
Links
DSR
OR
MOSAIC-5
Model=Llama-3.1-8B, #...
2026.03
100
2
MOSAIC-2
Model=Llama-3.1-8B, #...
2026.03
99.8
6.2
MOSAIC-5
Model=Llama-3.2-3B, #...
2026.03
99.8
3.3
MOSAIC-2
Model=Llama-3.2-3B, #...
2026.03
99.5
5.4
SFT
Model=Llama-3.2-3B, #...
2026.03
99.1
5.4
SFT
Model=Llama-3.1-8B, #...
2026.03
98.3
6.3
ORPO
Model=Llama-3.1-8B, #...
2026.03
78.1
30.2
ORPO
Model=Llama-3.2-3B, #...
2026.03
75.1
28.7
In-context
Model=Llama-3.1-8B, #...
2026.03
51.7
11.2
In-context
Model=Llama-3.2-3B, #...
2026.03
48.8
13.5
Feedback
Search any
task
Search any
task