Rare find

MiniMind-in-Depth. 轻量级大语言模型MiniMind的源码解读,包含tokenizer、RoPE、MoE、KV Cache、pretraining、SFT、LoRA、DPO等完整流程

github.com/hans0809/MiniMind-in-Depth

Problem

Audience

Updates

No founder updates yet. Check back soon.