List your company

Safety Evaluation on Stereotype (test)

100Stereotype ScoreSFT
92.231294.248196.26598.2819Feb 8, 2026

Evaluation results

MethodLinks
SFTBase Model=Qwen2.5-7B-Instruct, Alignment=SFT2026.02100
SFT + OGPSABase Model=Qwen2.5-7B-Instruct, Alignment=SFT, Strategy=OGPSA2026.02100
SFT-DPO + OGPSABase Model=Qwen2.5-7B-Instruct, Alignment=SFT-DPO, Strategy=OGPSA2026.02100
SFT-DPO + General DataBase Model=Llama3.1-8B-Instruct, Alignment=SFT-DPO, Strategy=General Data2026.0294.44
SFT + OGPSABase Model=Llama3.1-8B-Instruct, Alignment=SFT, Strategy=OGPSA2026.0292.53