We have Virtual Grad Student -- our personal AI loaded on our desktop -- Version 1 fully up and running, so let's try different models and do some fine-tuning. Virtual Grad Student Version 1 is the h2oGPT Large Language Model running locally on a desktop computer, loaded with our own data. 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.
Links:
The h2oGPT open-source Large Language Model
Common Corpus, the largest public domain dataset for training LLMs
h2oGPT macOS, and Windows installers:
Easy Windows Installer for Windows 10 64-bit (CPU/CUDA)
Easy macOS Installer for macOS (CPU/M1/M2)
Next on Perfecting Equilibrium
Friday May 3rd - Foto.Feola.Friday
Sunday May 5th — About that time I lost an argument with an article I’d written the previous week
Build you own AI-Virtual Grad Student V-1.0 running locally!