This research continues many of the directions our team worked on prior to Anthropic, including: GPT-3, Circuit-Based Interpretability, Multimodal Neurons, Scaling Laws, AI & Compute, Concrete Problems in AI Safety, and Learning from Human Preferences. Drive commercial negotiations and Master Purchase Agreements for critical electrical and mechanical infrastructure such as generators, medium- and low-voltage switchgear, transformers, UPS systems, chillers, cooling towers, CRAHs, and pumps.