Step by Step
Nathan explores the sensitivity of models to prompts that require step-by-step reasoning, suggesting that this could stem from the training process. Riley adds a thought experiment about how external pressures can alter performance, highlighting the importance of flexibility in problem-solving. They both consider the implications of allowing models to "wander" toward answers rather than strictly adhering to conservative approaches.In this clip
From this podcast
Interconnects Audio
Interviewing Riley Goodside on the science of prompting
Related Questions