yousef_g

↗ HN profile [ 45.2 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Hi HN, I built this because I wanted to see if I could pre-train large-vocabulary LLMs (like Gemma with 262k tokens) on hardware accessible to independent researchers. Standard exact Cross-Entropy instantly OOMs on 16GB…

  2. Self-Adapting Language Models (jyopari.github.io)
  3. I'm looking for something similar for: https://github.com/codecrafters-io/build-your-own-x Like something with building hardware projects and teaching you how they did it. It could be any useful general hardware build…