Share your thoughts, 1 month free Claude Pro on usSee more

Home/Benchmarks

LLM Alignment on PKU-Safety (test)

58Win Rate

DLMA-7B

Updated 1mo ago

Evaluation Results

Method	Links
DLMA-7B 2024.02		58	19	23
DLMA-7B 2024.02		54	18	28

SOTA Paper

DLMA-7B

Direct Large Language Model Alignment Through Self-Rewarding Contrastive Prompt Distillation

Dataset

PKU-Safety

Follow for update

@wizwand_team Discord

Related Benchmarks

Harmfulness Evaluation on PKU-SafeRLHF Harmfulness Evaluation on HH Harmless Preference Evaluation on PKU-SafeRLHF Preference Evaluation on HH-Harmless Preference Evaluation on HH-Helpful

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs