asuramaya / heinrich Star 3 Code Issues Pull requests Model forensics through geometry: residual-stream projections, activation traces, and attention routing. threejs pytorch ai-safety interpretability activation-analysis llm mechanistic-interpretability residual-stream model-forensics Updated Aug 26, 2026 Python
Emrevrg / ox-alpha-model-forensics Star 1 Code Issues Pull requests Independent model forensics investigation into OpenRouter's stealth ox-alpha model. machine-learning ai reverse-engineering artificial-intelligence glm multimodal ai-research large-language-models llm openrouter zhipu-ai model-forensics ox-alpha Updated Aug 23, 2026
dangjacob101 / model-eval-behavior-change Star 0 Code Issues Pull requests Testing which evaluation-context cues change Qwen3-32B's task-completion behavior. model-evaluation ai-safety qwen3 model-forensics evaluation-awareness Updated Aug 31, 2026 Python
justinbrianhwang / fl-datafree-backdoor-detection Star 0 Code Issues Pull requests Data-free backdoor detection in federated learning via reverse engineering (H100 experiments). adversarial-machine-learning federated-learning machine-learning-security backdoor-attacks data-free backdoor-attack model-forensics Updated Feb 12, 2026 Jupyter Notebook