from Hacker News

NanoMoE: Mixture-of-Experts (Moe) LLMs from Scratch in PyTorch

by danboarder on 4/6/25, 3:15 AM with 0 comments