Deep Dive Into LLMs
Master the architectural intricacies of transformers and attention mechanisms to design, train, and innovate cutting-edge generative models for advanced data science research. * **Master transformer architectures** and attention mechanisms from scratch * **Implement custom training loops** for fine-tuning large models * **Optimize model inference** using quantization and pruning techniques * **Architect novel generative models** for complex data science tasks
5 sections ยท 20 lessons
Course outline
Training Pipeline
- Three Training Stages
- Internet Data Curation
- Tokenization Fundamentals
- Neural Network Training
- Base Models Explained
- Text Generation Process
Post-Training
- Supervised Fine-Tuning
- Conversation Tokenization
Limitations
- Hallucinations
- Tool Use Integration
- Computational Limits
- Tokenization Blind Spots
Reinforcement Learning
- Reinforcement Learning Basics
- RL in Math & Code
- Chain of Thought
- RLHF for Subjective Tasks
- RLHF Limitations
Practical Insights
- Mental Model of AI
- LLM Ecosystem
- Future Capabilities