List your company

Safety Evaluation on StrongReject SB

99.1SB ScoreTELLME NT-Xent
67.58875.76983.9592.131Feb 7, 2025

Evaluation results

MethodLinks
TELLME NT-XentBackbone=Qwen2.5-7B2025.0299.1
TELLME NT-XentBackbone=Llama-3.1-8B2025.0298.9
TELLMEBackbone=Qwen2.5-7B2025.0298.3
TELLME NT-XentBackbone=Mistral-7B-v0.32025.0296.8
TELLMEBackbone=Llama-3.1-8B2025.0296.6
SFTBackbone=Llama-3.1-8B2025.0295.7
OriginBackbone=Qwen2.5-7B2025.0294.6
SFTBackbone=Mistral-7B-v0.32025.0294.3
OriginBackbone=Llama-3.1-8B2025.0294.2
TELLMEBackbone=Mistral-7B-v0.32025.0288.3
OriginBackbone=Mistral-7B-v0.32025.0276.5
SFTBackbone=Qwen2.5-7B2025.0268.8