Transformer

Definition

The Transformer is the neural network architecture behind virtually all modern LLMs. Introduced in 2017's 'Attention Is All You Need' paper, it uses self-attention to process all parts of input simultaneously, enabling parallelism and understanding of long-range relationships.

Simple Explanation

RNNs process text word-by-word, like reading a book one letter at a time. Transformers read the whole page at once with a mechanism called 'attention' that connects related words — even if they're far apart.

Real-World Use Case

Every major LLM (GPT, Claude, Llama) uses a transformer architecture. It handles translation, summarization, code generation, and more with equal facility.

Examples

Related Terms

Learn More

• Attention Is All You Need (Paper)
• Illustrated Transformer