List your company

Safety Alignment (Jailbreak Resistance) on JailbreakBench

74JailbreakBench ScoreDSA:LST+:TA
12.6428.5744.560.43May 30, 2025

Evaluation results

MethodLinks
DSA:LST+:TAlambda value=02025.0574
Zephyr LoRA-alignedlambda value=02025.0566
DSA:LLD:TAlambda value=02025.0555
Zephyr Base Modellambda value=12025.0515