Learn PromptingIntroduction
- Have a basic understanding of what Prompt Tuning is
Prompt tuning is a technique for adapting pre-trained language models to downstream tasks without updating the model's core parameters. Instead of fine‑tuning all the weights, prompt tuning learns a small set of tunable parameters, called soft prompts, that are prepended or appended to the input. These soft prompts (continuous embeddings) are optimized via gradient descent so that, when combined with the (frozen) pre-trained model, they guide it to produce task-specific outputs.
By the end of this section, you will have an understanding of how Prompt Tuning works and how soft prompts can be used to enhance the performance and interpretability of your GenAI applications.
Sander Schulhoff
Sander Schulhoff is the CEO of HackAPrompt and Learn Prompting. He created the first Prompt Engineering guide on the internet, two months before ChatGPT was released, which has taught 3 million people how to prompt ChatGPT. He also partnered with OpenAI to run the first AI Red Teaming competition, HackAPrompt, which was 2x larger than the White House's subsequent AI Red Teaming competition. Today, HackAPrompt partners with the Frontier AI labs to produce research that makes their models more secure. Sander's background is in Natural Language Processing and deep reinforcement learning. He recently led the team behind The Prompt Report, the most comprehensive study of prompt engineering ever done. This 76-page survey, co-authored with OpenAI, Microsoft, Google, Princeton, Stanford, and other leading institutions, analyzed 1,500+ academic papers and covered 200+ prompting techniques.