Machine learning at scale 🤖

LLM Training

Dive into the complete journey of training large language models (LLMs), from preparing datasets to the latest techniques for reinforcement learning and real-world examples from top companies. Whether you're building from scratch or refining your models, this guide will provide practical insights.

Whether you're a beginner or looking to master advanced LLM techniques, this guide covers practical code examples and hands-on strategies.

Explore the table of contents below to navigate through essential training topics.

Table of Contents

I. Fundamentals

  1. Introduction
  2. Dataset Preparation
    1. Create Alignment Data from Scratch
    2. Dedup Training Data
  3. Pretraining
  4. SFT and PEFT

II. Finetuning with RLHF

  1. PPO
  2. DPO
  3. IPO
  4. KTO
  5. ORTO
  6. CGPO
  7. Self Rewarding LLM
  8. REINFORCE
  9. Negative Preference Optimization (Unlearning)

III. Case Studies from Leading Companies

  1. OpenAI
  2. Google
  3. Apple
  4. NVIDIA
  5. Meta