Large Language Models
Overview
In this part, we examine how pre-training builds general language capabilities and how further training can adapt a model to tasks, instructions, and human preferences. We then outline the rise of GPT-style models, distinguish open-weight and open-source releases, and discuss developments in large language models.
The chapters in this part are as follows.
- Pre-training and Fine-tuning introduces the concepts of pre-training and fine-tuning, which are key to the development of large language models.
- Instruction Tuning and RLHF discusses two techniques that are used to align large language models with human intentions.
- Rise of GPT and Open-Weight Models outlines the rise of GPT-style models and publicly available model weights, including the distinction between open-weight and open-source releases.
- Recent Trends outlines some of the recent trends in large language models.
- Summary summarizes the key takeaways from this part.