Mental Models for Advanced ChatGPT Prompting with Riley Goodside

Today we’re joined by Riley Goodside, staff prompt engineer at Scale AI. In our conversation with Riley, we explore LLM capabilities and limitations, prompt engineering, and the mental models required to apply advanced prompting techniques. We dive deep into understanding LLM behavior, comparing k-shot and zero-shot prompting, and discussing major mental models such as the role of prompting, understanding RLHF, and the mechanism of autoregressive inference. We discuss Riley’s approach to prompting, addressing language model concerns as well as emphasizing the idea that prompting is a scaffolding structure that leverages the model context, resulting in achieving the desired model behavior and response rather than focusing solely on writing ability skills. The complete show notes for this episode can be found at twimlai.com/go/652.