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. Experience at hyperscalers, large compute consumers, or data center operators with exposure to infrastructure lease structures (including colocation, dark fiber IRUs, and compute / equipment leases).