Chapter 4 · GPT Model

Chapter 4 assembles the pieces — normalization, feed-forward, and attention — into a transformer block, then stacks blocks into the full GPT.

Lessons

1. LayerNorm & GELU feed-forward
2. The transformer block
3. The full GPT model

Code

• @node-llm/core → src/model/block.ts, src/model/gpt.ts (GPT_CONFIG_124M)
• Runnable: pnpm ch04