Chapters (DL4NLP 2)

This chapter will review the architectural components of the Transformer, which is (in some variant) still the backbone of most contemporary large language models. For students taking this course, it is crucial to be familiar with these basics to be able to follow along.

tbd

BERT-style fine-tuning vs. SFT for gen LLMs vs. Prompt-based Learning Adapters LoRa / QLoRa Contrastive Finetuning of Embedding Models

Additional Resources Overview of GPT-1 until GPT-3