I Understand Most of the Words I'm Saying: An Introduction to How LLMs Work
An introduction to a six-part series on how large language models work, from a network engineer who understands most of it and a little of the math.
# series
A six-part look at what's inside a large language model, how it gets trained, and what it runs on, written by a network engineer for people who would rather know how the thing works than be told it's magic.
browse: /categories /series /tags /archive
5 of 6 published
An introduction to a six-part series on how large language models work, from a network engineer who understands most of it and a little of the math.
[ Ai and llms ] Part 1 of 6 11 min read
How LLMs Actually WorkA large language model is an enormous pile of numbers, and everything it does comes from multiplying other numbers by that pile. Part one of six.
[ Ai and llms ] Part 2 of 6 9 min read
How LLMs Actually WorkTraining turns roughly 405 billion random numbers into something useful through one boring move: guess the next token, get graded, nudge every number a hair, and repeat. Part two of six.
A frontier training run is a networking problem dressed up in a math costume. The industry chose the chickens over the oxen, so how do you get 16,000 of them pulling in the same direction? Part three of six.
Training is one enormous bill, paid once. Inference is the meter that never stops running, and an old network saying about bandwidth and latency explains why a chatbot's words arrive one at a time. Part four of six.
[ Ai and llms ] Part 5 of 6 10 min read
How LLMs Actually WorkA language model has no memory in the sense you'd expect. Nearly everything that looks like memory is text being resent, summarized, searched, or stored by the software around it, and the diary gets heavy fast. Part five of six.