IV — Transformers & LLMs

This part covers the architecture behind modern NLP — the Transformer — and how pretrained language models are applied to downstream tasks.

Chapters in this part

• Chapter 4: Transformers & LLMs — Attention, the Transformer architecture, and working with pretrained models. Paired with an illustrated, step-by-step walkthrough.

What you'll be able to do

• Explain self-attention and the Transformer block.
• Build sequence-to-sequence and attention-based models for machine translation.
• Use pretrained Transformer models (e.g. BERT) for downstream tasks.