centerforaisafety / wmdpLinks
WMDP is a LLM proxy benchmark for hazardous knowledge in bio, cyber, and chemical security. We also release code for RMU, an unlearning method which reduces LLM performance on WMDP while retaining general capabilities.
☆157Updated 7 months ago
Alternatives and similar repositories for wmdp
Users that are interested in wmdp are comparing it to the libraries listed below
Sorting:
- Improving Alignment and Robustness with Circuit Breakers☆255Updated last year
- Röttger et al. (NAACL 2024): "XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models"☆123Updated 11 months ago
- ☆192Updated 2 years ago
- [ICLR 2025] Official Repository for "Tamper-Resistant Safeguards for Open-Weight LLMs"☆66Updated 7 months ago
- [ICML 2024] COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability☆177Updated last year
- Official implementation of AdvPrompter https//arxiv.org/abs/2404.16873☆174Updated last year
- A Comprehensive Assessment of Trustworthiness in GPT Models☆311Updated last year
- Benchmark evaluation code for "SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal" (ICLR 2025)☆74Updated 10 months ago
- ICLR2024 Paper. Showing properties of safety tuning and exaggerated safety.☆92Updated last year
- ☆60Updated 2 years ago
- Official implementation of ICLR'24 paper, "Curiosity-driven Red Teaming for Large Language Models" (https://openreview.net/pdf?id=4KqkizX…☆87Updated last year
- [ICLR 2025] General-purpose activation steering library☆137Updated 4 months ago
- Official Repository for ACL 2024 Paper SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding☆152Updated last year
- Finding trojans in aligned LLMs. Official repository for the competition hosted at SaTML 2024.☆116Updated last year
- Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses (NeurIPS 2024)☆65Updated last year
- [ICML 2025] Weak-to-Strong Jailbreaking on Large Language Models☆92Updated 8 months ago
- This is the starter kit for the Trojan Detection Challenge 2023 (LLM Edition), a NeurIPS 2023 competition.☆90Updated last year
- We jailbreak GPT-3.5 Turbo’s safety guardrails by fine-tuning it on only 10 adversarially designed examples, at a cost of less than $0.20…☆336Updated last year
- Stanford NLP Python library for benchmarking the utility of LLM interpretability methods☆162Updated 7 months ago
- Repo for the research paper "SecAlign: Defending Against Prompt Injection with Preference Optimization"☆82Updated 6 months ago
- A resource repository for representation engineering in large language models☆148Updated last year
- A re-implementation of the "Red Teaming Language Models with Language Models" paper by Perez et al., 2022☆35Updated 2 years ago
- ☆27Updated last year
- LLM experiments done during SERI MATS - focusing on activation steering / interpreting activation spaces☆100Updated 2 years ago
- Package to optimize Adversarial Attacks against (Large) Language Models with Varied Objectives☆70Updated last year
- Codes and datasets of the paper Red-Teaming Large Language Models using Chain of Utterances for Safety-Alignment☆108Updated last year
- A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and Toxicity.☆85Updated 10 months ago
- [ICLR'24] RAIN: Your Language Models Can Align Themselves without Finetuning☆98Updated last year
- Python package for measuring memorization in LLMs.☆179Updated 6 months ago
- A lightweight library for large laguage model (LLM) jailbreaking defense.☆60Updated 4 months ago