How do large language models work?
#How to work with large language models
visual kraken
thin emberBOT
Hi @visual kraken. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
Large language models, such as OpenAI's GPT-3, are functions that map text to text. Given an input string of text, a large language model predicts the text that should come next. These models are trained to minimize prediction error over vast quantities of text, and through this process, they end up learning concepts useful for these predictions. For example, they learn how to spell, how grammar works, how to paraphrase, how to answer questions, how to hold a conversation, how to write in many languages, how to code, and more. None of these capabilities are explicitly programmed in—they all emerge as a result of training. GPT-3 powers hundreds of software products, including productivity apps, education apps, games, and more.
To control a large language model, the most influential input is the text prompt. Large language models can be prompted to produce output in a few ways:
- Instruction: Tell the model what you want.
- Completion: Induce the model to complete the beginning of what you want.
- Demonstration: Show the model what you want, with either a few examples in the prompt or many hundreds or thousands of examples in a fine-tuning training dataset.
For example, instruction-following models (e.g., text-davinci-003 or any model beginning with text-) are specially designed to follow instructions. Write your instruction at the top of the prompt (or at the bottom, or both), and the model will do its best to follow the instruction and then stop. Instructions can be detailed, so don't be afraid to write a paragraph explicitly detailing the output you want.
You can find more information on how to work with large language models in the OpenAI Cookbook (https://github.com/openai/openai-cookbook/blob/main/articles/how_to_work_with_large_language_models.md).
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
visual kraken
@thin ember tell me all about Code Capabilities
thin emberBOT
Let me check the knowledge base for you. One moment! :books:
Large language models are not only proficient at handling text, but they can also be highly effective with code. OpenAI has developed a specialized code model called Codex. This model powers more than 70 products, including:
- GitHub Copilot (https://copilot.github.com/): This tool autocompletes code in VS Code and other IDEs.
- Pygma (https://pygma.app/): Pygma turns Figma designs into code.
- Replit (https://replit.com/): Replit features an 'Explain code' button among other features.
- Warp (https://www.warp.dev/): Warp is a smart terminal with AI command search.
- Machinet (https://machinet.net/): Machinet writes Java unit test templates.
It's important to note that unlike instruction-following text models (e.g., text-davinci-002), Codex is not trained to follow instructions. As a result, designing good prompts for Codex can require more care.
You can find more information about the code capabilities of large language models in the OpenAI Cookbook (https://github.com/openai/openai-cookbook/blob/main/articles/how_to_work_with_large_language_models.md).
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
visual kraken
(How often is kapa updated?)
void vale
@visual kraken just realized I never got back to you here! Kapa is kept up to date minimum once per week. But we usually refresh every few days (esp. if major new releases happen). 
thin emberBOT
Let me check the knowledge base for you. One moment! :books: