Synthetic Persona Pretraining: Alignment from Token Zero 15 August 2026·10 mins Machine Learning Paper Notes Machine-Learning Arxiv Alignment Pretraining Llm Safety
Understanding the Inner Thoughts of AI — DeepMind on Interpretability 11 July 2026·5 mins Machine Learning Talk Notes Llm Interpretability Alignment Deepmind Mechanistic-Interpretability Sparse-Autoencoders