Running your own AI on your desktop
Chris Feola
Author, Perfecting Equilibrium: For a brief, shining moment Web1 democratized data. Then Web2 came along and made George Orwell look like an optimist. Now Web3 is Perfecting John Nash’s Information Equilibrium.
First, sorry about my giant head! As you’ll hear, I had to reshoot this video several times, and ran out of time. My wallpaper is a picture of me and my grandkids wearing kewl sunglasses together; the grandkids are covered up, so my daughter won’t murder me, thankfully. Anyway this week we're installing and running the h2oGPT Large Language Model on a desktop computer to show how you run your own AI locally. I’ve gone on and on about how Large Language Models — so-called AIs — are crippled by training on bad data. Garbage In; Garbage Out! So let’s build a local LLM and feed it good data for a specific purpose. Welcome to Virtual Grad Student! We’re going to set up a Large Language Model to run locally, feed it a clean set of data, then make it available to authors as a virtual writer’s assistant. For example, to pull together a few paragraphs of background on Roman aqueduct architecture. Oh, and sorry again about the background!
Links:
Common Corpus, the largest public domain dataset for training LLMs
h2oGPT macOS, and Windows installers:
Next on Perfecting Equilibrium
Friday April 19th?- Foto.Feola.Friday
Sunday April 21st — There will be no 2nd Civil War The US Civil War was the first great conflict of the Industrial Age, won in large part by the manufacturing might of the Union. Such wars have faded away along with the Industrial Age. Today’s conflicts are being fought with cheap homemade drones and cyberattacks and electronic trade. If there’s a Third Battle of Bull Run, only electrons will see it.