Transformers: The Definitive Guide

Table of contents

1. Brief Table of Contents (Not Yet Final)
2. 1. Intro to Transformers and Attention
   1. Transformer basics
      1. Tokenizer: Text representation in the transformer
      2. Token and positional embeddings
      3. Attention mechanism
      4. Encoder and decoder parts
   2. Enhancements in transformer design: Longer context and attention variations
      1. Longer context windows with better performance
      2. Attention mechanism variations
      3. Conclusion
3. 2. Leveraging and Refining LLMs
   1. Using instruction models for common NLP tasks
      1. Accessing the LLaMA 3 instruct model
      2. Text Generation
      3. Text Summarization
      4. Named Entity Recognition
      5. Text Classification
   2. Advanced prompting techniques
      1. Chain-of-thought prompting
      2. Zero-shot CoT prompting
      3. Tree of thought prompting
      4. Thread of thought prompting
   3. Aligning your model with your preferences
      1. Supervised Fine-Tuning
      2. Reinforcement learning from human feedback
      3. Direct preference optimization
      4. Odds Ration Performance Optimization
      5. Conclusion
4. About the Author