from-scratch

When Anthropic announced it would watermark text output from Claude models, Sebastian Raschka posted a short explanation of the mechanism. The post went unexpectedly viral — not because watermarking is exciting, but because almost nobody could say concretely what it does. He planned a ten-slide follow-up. It became fifty. The result is one of the better explainers of the year, and it doubles as a clean walkthrough of how LLM sampling actually works....

Weekly Video Notes — a short article distilling one talk from the weekly digest. Source video and key frames are embedded throughout. 📜 Classic of the Week. This week’s digest re-surfaces a 2023 classic: Andrej Karpathy’s “Let’s build GPT: from scratch, in code, spelled out.” Three years later it is still, hands down, the clearest single resource for understanding what is actually happening inside a transformer. If you have never sat with it end-to-end, this is your nudge....