1 weeks ago
From Transformer to GPT
Part 8 of the AI/LLM mastery series — the capstone of the architecture. How a stack of Transformer blocks becomes a working GPT: d…
AI/LLM Security → Transformers & Architecture
Part 8 of the AI/LLM mastery series — the capstone of the architecture. How a stack of Transformer blocks becomes a working GPT: d…
Part 7 of the AI/LLM mastery series. One self-attention is not a Transformer — this assembles the full block: positional encoding …
Part 6 of the AI/LLM mastery series — the math of attention, worked by hand. Query, key and value vectors; the dot-product score; …
Part 5 of the AI/LLM mastery series. Language is a sequence — order matters, words depend on far-away words, and lengths vary — wh…