LLMs & Transformers: Dive into Hugging Face, Fine-tuning, Tokenization, and Datasets
What you will learn
How to build your own tokenizer using Hugging Face’s Transformers library
How to build a custom dataset in Hugging Face
How to train a GPT-2 model from scratch using Hugging Face’s Transformers library
How to instruction fine-tune a LLM using PEFT
Description
Welcome to “Mastering Generative AI with Large Language Models: A Hugging Face Guide”. In today’s AI-driven world, Large Language Models (LLMs) have revolutionized the realm of generative AI, enabling machines to generate human-like text, answer questions, and even author original content. This course is meticulously crafted to provide you with a deep understanding of these models and how to harness their power using the renowned Hugging Face platform.
Our journey begins with a robust introduction to the world of LLMs, deciphering their intricacies, and exploring the management of their compute requirements. From there, we transition into the Hugging Face universe, a pivotal platform offering an array of pre-trained models ready to be utilized in innovative applications.
But understanding the theory isn’t enough; practical knowledge is vital. This is why our second section delves deep into the workings of Transformers, one of the core components behind LLMs. Get hands-on with manipulating datasets, building your custom models, and understanding the art of tokenization.
Lastly, a special emphasis is laid on training and fine-tuning. Learn how to tailor LLMs to your specific needs, be it summarization or text generation. With techniques like Instruction Fine-tuning and PEFT, you’ll master the art of tweaking models to perfection. We will finish off the course by training a GPT-2 completely from scratch to generate text on our custom dataset.
Content