Chess as autocomplete
Can autocomplete play chess blindfolded? A journey into applied AI research
Training a transformer from scratch to play blindfold chess, with the same recipe used to train modern LLMs.
Blog
Do you even know how LLMs work?
What they are, how they work, and how they are trained.
This post explores LLMs, the models behind services like ChatGPT, Claude, and Gemini. The goal is to give you an in depth but approachable understanding of LLMs, how they work, and how they are trained. We will ignore some details, and provide simplified explanations where appropriate.