Hands-On Large Language Models

AI has gained remarkable new language capabilities in recent years. Driven by rapid advancements in deep learning, language AI systems are now able to write and understand text better than ever before. This trend is enabling the emergence of new features, products, and entire industries. Through the visually educational nature of this book, Python developers will learn the practical tools and concepts they need to leverage these capabilities today.

You will learn how to harness the power of pre-trained large language models for use cases such as text generation and summarization; create semantic search systems that go beyond keyword search; build systems that classify and cluster texts to enable scalable understanding of a large number of text documents; and utilize existing libraries as well as pre-trained models for text classification, search, and clustering.

This book will also show you how to:

• Create advanced LLM pipelines to cluster text documents and explore the topics they belong to
• Develop semantic search engines that go beyond keyword search, using methods like dense retrieval and re-rankers
• Discover various use cases where these models can provide added value
• Understand the architecture of the underlying transformer models such as BERT and GPT
• Gain a deeper understanding of how LLMs are trained
• Optimize LLMs for specific applications using methods like fine-tuning generative models, contrastive fine-tuning, and in-context learning

Jay Alammar is the Director and Engineering Fellow at Cohere, a pioneering provider of large language models as an API. Maarten Grootendorst is a Senior Clinical Data Scientist at the Netherlands Comprehensive Cancer Organisation.