"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis cover image

Delving into The Prompt Report, with Sander Schulhoff of LearnPrompting.org

"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis

NOTE

Continuous Improvement Through Fine-Tuning and Edge Cases Coverage

Fine-tuning on chain of thought reasoning, progressively covering edge cases over time, and including failed examples in the fine-tuning dataset seem to show the most significant improvements. While this approach may be cost and latency prohibitive, it appears to be more effective than trying to cover all edge cases in a single prompt. Despite the success, there remains uncertainty about its effectiveness and the decision-making process involved.

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner