New comment by rpdaiml in "Show HN: I built a tiny LLM to demystify how language models work"

rpdaiml — Mon, 06 Apr 2026 14:44:35 +0000

This is a nice idea. A tiny implementation can be way more useful for learning than yet another wrapper around a big model, especially if it keeps the training loop and inference path small enough to read end to end.

New comment by rpdaiml in "Show HN: 1-Bit Bonsai, the First Commercially Viable 1-Bit LLMs"

rpdaiml — Thu, 02 Apr 2026 15:10:11 +0000

curious and wondering its porting and usage on edge devices?

Hacker News: rpdaiml

New comment by rpdaiml in "Show HN: I built a tiny LLM to demystify how language models work"

New comment by rpdaiml in "Show HN: 1-Bit Bonsai, the First Commercially Viable 1-Bit LLMs"