Quick Start Guide to Large Language Models: Strategies and Best Practices for Using ChatGPT and Other LLMs

4

Optimizing LLMs with Customized Fine-Tuning

Introduction

So far, we’ve exclusively used LLMs, both open- and closed-source, just as they are off the shelf. We were relying on the power of the Transformer’s attention mechanisms and their speed of computation to perform some pretty complex problems with relative ease. As you can probably guess, that isn’t always enough.

In this chapter, we will delve into the world of fine-tuning large language models (LLMs) to unlock their full potential. Fine-tuning updates off-the-shelf models and empowers them to achieve higher-quality results; it can lead to token savings, and often lower-latency requests. While GPT-like LLMs’ pre-training on extensive text data enables impressive few-shot learning capabilities, ...