Last year we trained GPT-3 and made it available in our API. With only a few examples, GPT-3 can perform a wide variety of natural language tasks, a concept called few-shot learning or prompt design. Customizing GPT-3 can yield even better results because you can provide many more examples than what’s possible with prompt design.

You can customize GPT-3 for your application with one command and use it immediately in our API:

It takes less than 100 examples to start seeing the benefits of fine-tuning GPT-3 and performance continues to improve as you add more data. In research published last June, we showed how fine-tuning with less than 100 examples can improve GPT-3’s performance on certain tasks. We’ve also found that each doubling of the number of examples tends to improve quality linearly.

With one of our most challenging research datasets, Grade School Math problems, fine-tuning GPT-3 improves accuracy by 2 to 4x over what’s possible with prompt design.

// `components` object you'll pass to PortableText w/ optional TS definition
import {PortableTextComponents} from '@portabletext/react'

const components: PortableTextComponents = {
  marks: {
    // Ex. 1: custom renderer for the em / italics decorator
    em: ({children}) => <em className="text-gray-600 font-semibold">{children}</em>,

    // Ex. 2: rendering a custom `link` annotation
    link: ({value, children}) => {
      const target = (value?.href || '').startsWith('http') ? '_blank' : undefined
      return (
        <a href={value?.href} target={target} rel={target === '_blank' && 'noindex nofollow'}>
          {children}
        </a>
      )
    },
  },
}