From the course: Prompt Engineering with LangChain

Unlock the full course today

Join today to access over 24,600 courses taught by industry experts.

Zero-shot prompts

Zero-shot prompts

From the course: Prompt Engineering with LangChain

Start my 1-month free trial Buy for my team

Zero-shot prompts

“

- [Instructor] Because large language models were trained on such a vast amount of data, they possess a surprising feature. And that feature is the ability to do what's called zero shot learning. And Zero Shot learning facilitates something called zero shot prompting. So what is Zero Shot prompting? Well, it involves asking the model to perform a task without giving it any specific examples beforehand. This is going to allow the language model to generate responses to a prompt without requiring specific examples or even being specifically trained for that task. That's because of the way language models were pre-trained. They're trained on vast quantities of data, and because of this, again, they're able to execute tasks with what's called the zero shot. For example, to classify a text as neutral, negative, or positive, you might prompt the language model by saying, classify the text into neutral, negative, or positive. And the model could respond with, for example, neutral for a…

Contents