Skip to content

Andrew Fairless, Ph.D.

Data, Science, and Tinkering

Overview
Experience and Education
Publications
SHAP Tutorial
Understanding the Basics of Bayesian Linear Regression
Classifying Medicine
The Peanuts Project

Search for:

Search for:

What I Watch: How LLMs store facts

Home/What I Learn/What I Watch: How LLMs store facts

By BylineAndrew Fairless on December 16, 2024September 3, 2024

How might LLMs store facts | Chapter 7, Deep Learning
3Blue1Brown
Aug 31, 2024

“Unpacking the multilayer perceptrons in a transformer, and how they may store facts”

Cat Links What I Learn Tag Links attention large language model linear algebra machine learning natural language processing neural network transformer

Post navigation

What I Watch: compare high dimensional vectorsPrev post

What I Read: LLM Agents in ProductionNext post

Categories

Bayesian statistics Machine Learning Statistics What I Learn What I Make

Tags

artificial intelligence attention Bayesian chatbot cognition computer vision database data engineering data science deployment DevOps economics efficiency embedding generalization generative GPU graph healthcare image interpretability large language model latency linear algebra machine learning medicine MLOps monitoring natural language processing neural network neuroscience optimization pipeline probability Python recurrent regression reinforcement learning scalability software engineering SQL statistics training transformer unit test

Copyright © 2024 Andrew Fairless, Ph.D.. All Rights Reserved. | Simple Persona by Catch Themes

Scroll Up