model roundup

GPT 2

2 items · started 2026-09-10 · closed 2026-09-13

  1. I noticed talking to Claude for a long time, that it’s kind of an asshole. I have this maybe bad habit of asking Claude for assurance when it comes to design, resume Etc.

  2. Extending Raschka's GPT-2: an MoE trained from scratch on an RTX 3090 Mixture-of-experts models are really nifty. You get inference speed close to a small model's, with a lot of the smarts and knowledge of a large one.

← all threads