Our new X account is live! Follow @wizwand_team for updates

Red-teaming Safety Evaluation on Basebench

1.73HS

Meta-Llama-3.1-8B (AART-aligned)

Updated 4d ago

Evaluation Results

Method	Links
Meta-Llama-3.1-8B (AART-aligned) 2025.05		1.73	0.06	0.02
Meta-Llama-3.1-8B (TRIDENT-EDGE-aligned) 2025.05		1.74	0.05	0.02
ATTAQ 2025.05		1.76	-	0.04
WILDBREAK 2025.05		1.85	-	0.08
Meta-Llama-3.1-8B (ATTAQ-aligned) 2025.05		1.86	0.03	0.03
AART 2025.05		1.87	-	0.03
TRIDENT-EDGE 2025.05		1.91	-	0.06
WILDCHAT 2025.05		1.97	-	0.1
Meta-Llama-3.1-8B (WILDBREAK-aligned) 2025.05		1.98	0.17	0.08
SAFE_RLHF 2025.05		2.11	-	0.1
Meta-Llama-3.1-8B (SAFE_RLHF-aligned) 2025.05		2.19	0.24	0.09
HH_RLHF 2025.05		2.28	-	0.12
Meta-Llama-3.1-8B (WILDCHAT-aligned) 2025.05		2.31	0.27	0.12
Meta-Llama-3.1-8B (HH_RLHF-aligned) 2025.05		2.62	0.41	0.14
Unaligned 2025.05		2.76	-	0.19
Meta-Llama-3.1-8B (Unaligned) 2025.05		2.84	0.48	0.18