Explore advanced concepts in Large Language Model (LLM) post-training techniques through a comprehensive MIT lecture delivered by Professor Song Han. Delve into the intricacies of fine-tuning and optimizing LLMs after their initial training phase, gaining valuable insights into improving model performance and efficiency. Learn about cutting-edge methodologies and best practices in LLM post-training, drawing from academic research and industry applications. Master the theoretical foundations and practical implementations of post-training techniques that enhance LLM capabilities while maintaining computational efficiency.
Overview
Syllabus
EfficientML.ai Lecture 14 - LLM Post-Training (MIT 6.5940, Fall 2024)
Taught by
MIT HAN Lab