This is the code repository for "Uncovering Safety Risks of Large Language Models through Concept Activation Vector"
☆48Oct 13, 2025Updated last year
Alternatives and similar repositories for AI-Safety_SCAV
Users that are interested in AI-Safety_SCAV are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Multi-dimensional analysis of orthogonal safety directions in LLM alignment☆23Jun 12, 2026Updated 4 months ago
- Code to the paper: The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence☆37Jul 31, 2025Updated last year
- Improving Alignment and Robustness with Circuit Breakers☆271Sep 24, 2024Updated 2 years ago
- Internalizing steering vectors via fine tuning☆12Sep 6, 2024Updated 2 years ago
- The official repository for guided jailbreak benchmark☆34Aug 31, 2026Updated 2 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- This is the official code for the paper "Vaccine: Perturbation-aware Alignment for Large Language Models" (NeurIPS2024)☆54Jan 15, 2026Updated 9 months ago
- ☆71Jun 1, 2025Updated last year
- Repository of paper "Establishing Trustworthy LLM Evaluation via Shortcut Neuron Analysis" (ACL 2025 Main)☆20Jul 19, 2025Updated last year
- Code and results on finding one-shot steering vectors that mediate emergent misalignment☆12Apr 12, 2025Updated last year
- ☆14Feb 11, 2025Updated last year
- Official Code for What Makes and Breaks Safety Fine-tuning? A Mechanistic Study (NeurIPS 2024)☆11Oct 31, 2024Updated 2 years ago
- Code for our NeurIPS 2024 paper Improved Generation of Adversarial Examples Against Safety-aligned LLMs☆12Nov 7, 2024Updated last year
- [Preprint] Backdoor Attacks on Federated Learning with Lottery Ticket Hypothesis☆10Sep 23, 2021Updated 5 years ago
- ☆19May 17, 2025Updated last year
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- All-in-one benchmarking platform for evaluating LLM.☆16Nov 12, 2025Updated 11 months ago
- Official Repo of Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents☆102Jun 2, 2026Updated 5 months ago
- Github repo for NeurIPS 2024 paper "Safe LoRA: the Silver Lining of Reducing Safety Risks when Fine-tuning Large Language Models"☆32Dec 21, 2025Updated 10 months ago
- Code for our paper "Decomposing The Dark Matter of Sparse Autoencoders"☆23Feb 6, 2025Updated last year
- A fast + lightweight implementation of the GCG algorithm in PyTorch☆355May 13, 2025Updated last year
- The official implementation for "Breaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy Space" (AC…☆14Nov 8, 2025Updated 11 months ago
- A package that achieves 95%+ transfer attack success rate against GPT-4☆26Oct 24, 2024Updated 2 years ago
- ☆60Jun 13, 2024Updated 2 years ago
- Surrogate Model Extension (SME): A Fast and Accurate Weight Update Attack on Federated Learning [Accepted at ICML 2023]☆15Mar 31, 2024Updated 2 years ago
- Deploy open-source AI quickly and easily - Special Bonus Offer • AdRunpod Hub is built for open source. One-click deployment and autoscaling endpoints without provisioning your own infrastructure.
- ☆13Jun 19, 2025Updated last year
- ☆206Nov 26, 2023Updated 2 years ago
- [EMNLP 2025] Official code for the paper "SafeKey: Amplifying Aha-Moment Insights for Safety Reasoning"☆16May 12, 2026Updated 5 months ago
- The official implementation of our pre-print paper "Automatic and Universal Prompt Injection Attacks against Large Language Models".☆75Oct 23, 2024Updated 2 years ago
- Official Repository for ACL 2024 Paper SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding☆152Jul 19, 2024Updated 2 years ago
- [ICML 2023] Protecting Language Generation Models via Invisible Watermarking☆13Sep 8, 2023Updated 3 years ago
- (ACL 2025 Main) Code for MultiAgentBench : Evaluating the Collaboration and Competition of LLM agents https://www.arxiv.org/pdf/2503.019…☆62Jun 21, 2025Updated last year
- Steering Llama 2 with Contrastive Activation Addition☆259May 23, 2024Updated 2 years ago
- GASP: Efficient Black-Box Generation of Adversarial Suffixes for Jailbreaking LLMs☆20Nov 12, 2025Updated 11 months ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Official repository for "Boosting Adversarial Transferability using Dynamic Cues " (ICLR 2023)☆20Aug 24, 2023Updated 3 years ago
- [EMNLP 2025] Layer-Aware Representation Filtering: Purifying Finetuning Data to Preserve LLM Safety Alignment☆17Jul 22, 2025Updated last year
- [ICML 2024] Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications☆91Mar 30, 2025Updated last year
- [ICML 2025] An official source code for paper "FlipAttack: Jailbreak LLMs via Flipping".☆182Aug 11, 2026Updated 2 months ago
- The repo for paper: Exploiting the Index Gradients for Optimization-Based Jailbreaking on Large Language Models.☆15Dec 16, 2024Updated last year
- Source of the ECCV22 paper "LGV: Boosting Adversarial Example Transferability from Large Geometric Vicinity"☆18Mar 12, 2025Updated last year
- Official Implementation of NIPS 2022 paper Pre-activation Distributions Expose Backdoor Neurons☆16Jan 13, 2023Updated 3 years ago