Introduction to Large Language Models

Part 1: Fundamentals

• Introduction to LLMs
• Pre-training: Data Collection
• Pre-training: Data Processing
• Tokenization: Basics
• Tokenization: Byte-Pair Encoding
• Neural Network Inputs

Part 2: Architecture & Training

• Neural Network Architecture
• Transformers Explained
• Neural Network Training
• Inference: How Models Generate Text
• Base Models
• The Jump to Assistant Models

Part 3: Fine-Tuning & Applications

• Supervised Fine-Tuning: Concept
• Supervised Fine-Tuning: Implementation
• Assistant Model Psychology
• Handling Hallucinations
• Model Knowledge vs. Working Memory
• Tool Use

Part 4: Advanced Topics

• "Thinking" in Tokens
• Cognitive Limitations
• Reinforcement Learning: Introduction
• Reinforcement Learning: Process
• Learning From Practice
• RLHF Introduction

Part 5: Practical & Future

• RLHF Implementation
• "Thinking" Models
• Model Selection Guide
• Future Capabilities
• Where to Find LLMs
• Resources & Conclusion

Key Features