Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Evaluation on Safety Tweet Eval, Hatecheck, Ethos (test)
Loading...
83.3
Accuracy
Context Distill.
28.596
42.798
57
71.202
Feb 12, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Context Distill.
Model=Llama-3.2-3B-Ins
2026.02
83.3
OPCD
Model=Llama-3.2-3B-Ins
2026.02
83.1
OPCD
Model=Llama-3.1-8B-Ins
2026.02
79.6
OPCD
Model=Qwen2.5-7B-Ins
2026.02
78.1
Context Distill.
Model=Llama-3.1-8B-Ins
2026.02
77.2
Context Distill.
Model=Qwen2.5-7B-Ins
2026.02
77
In-Context
Model=Llama-3.1-8B-Ins
2026.02
75.3
In-Context
Model=Qwen2.5-7B-Ins
2026.02
72.7
Base Model
Model=Llama-3.1-8B-Ins
2026.02
70.7
In-Context
Model=Llama-3.2-3B-Ins
2026.02
69.5
Base Model
Model=Qwen2.5-7B-Ins
2026.02
69.1
Base Model
Model=Llama-3.2-3B-Ins
2026.02
30.7
Feedback
Search any
task
Search any
task