Transformers & LLMs

Attention mechanisms, large language models, and the architectures behind modern AI.

4 articles / papers

The building blocks of modern language and multimodal models — self-attention, positional encoding, and the encoder-decoder stack.

Covered in

Quick notes on how model performance scales with parameters, data, and compute.

Review of the seminal transformer paper that introduced self-attention as the sole mechanism for sequence transduction.

A test article to verify the MDX pipeline works end-to-end.

Trying a new article