Notes

Distilling a Bigram 12 Sep 2026
Build a Bigram Model in 10 Minutes 30 Aug 2026
Memory to Train a Transformer Draft · 11 Jul 2026
Estimating Layers in Transformers 27 Jun 2026
Why KV Cache and Not QKV Cache? 12 Jun 2026
Is RL Secretly Supervised Learning? 28 Dec 2025

Currently Reading — LLM-Generated Notes

Estimating the Compute and Memory Needed for Language Models Draft · 11 Jul 2026

Also: CEO Hour — a timed thinking practice you can run on your own.