Stories by tosh

How to Recover from Microservices

How do the 7B, 13B, 30B, and 65B parameter models compare?

Glicko Rating System

GPT-4 architecture: what we can deduce from research literature

Safari Technology Preview 169

Discord is making all users pick a new username

We are in a new era of acute risk

SparseGPT: Language Models Can Be Accurately Pruned in One-Shot

Building the New Tinygrad Computer

Jeremy Howard Mojo Demo

My SaaS makes $50K MRR and is built on vanilla Ruby on Rails and jQuery

Supine Computing

AI with Sam Altman: The End of the World? Or the Dawn of a New One?

Clownfish: Constrained Decoding for LLMs Against JSON Schema

The zeitgeist of the thought leaders does tend negative

Nvidia GPT-2B-001

Three space trailing comments (2014)

Dynamic Languages Wizards: Panel on Language Design (2001)

Semantic Kernel