Member-only story
A Practical Introduction to LLMs
3 levels of using LLMs in practice
This is the first article in a series on using Large Language Models (LLMs) in practice. Here I will give an introduction to LLMs and present 3 levels of working with them. Future articles will explore practical aspects of LLMs, such as how to use OpenAI’s public API, the Hugging Face Transformers Python library, how to fine-tune LLMs, and how to build an LLM from scratch.
What is an LLM?
LLM is short for Large Language Model, which is a recent innovation in AI and machine learning. This powerful new type of AI went viral in Dec 2022 with the release of ChatGPT.
For those enlightened enough to live outside the world of AI buzz and tech news cycles, ChatGPT is a chat interface that ran on an LLM called GPT-3 (now upgraded to either GPT-3.5 or GPT-4 at the time of writing this).
If you’ve used ChatGPT, it’s obvious that this is not your traditional chatbot from AOL Instant Messenger or your credit card’s customer care.
This one feels different.
