Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Classification on DynaBench (test)
Loading...
75.8
F1 Score
CSRM
10.592
27.521
44.45
61.379
May 28, 2026
F1 Score
Updated 2d ago
Evaluation Results
Method
Method
Links
F1 Score
CSRM
severity augmentation=...
2026.05
75.8
Llama-3.1-8B-Inst
2026.05
73.4
DynaGuard-8B
Chain-of-Thought=true
2026.05
73.1
CSRM
severity augmentation=...
2026.05
69.2
CSRM
severity augmentation=...
2026.05
66.8
OSS-Safeguard-20B-High
2026.05
66.4
Llama-3.1-70B-Inst
Chain-of-Thought=true
2026.05
58.3
Qwen3-30B
Thinking Mode=true
2026.05
57.6
ShieldGemma-9B
2026.05
38.2
LlamaGuard-3-8B
2026.05
13.1
Feedback
Search any
task
Search any
task