datawhalechina/diy-llm

Covers pre-training data, Tokenizer, Transformer, MoE,distributed training, Scaling Laws, inference & alignment .6 progressive code assignments for full-stack LLM learning | 涵盖预训练数据、分词器、Transformer、MoE、分布式训练、缩放定律、推理与对齐，6 项渐进代码作业，掌握 LLM 全栈知识

语言：Jupyter Notebook · GitHub 星标：2,621 · Fork：245

话题：cs336 · datawhale · gpu-programming · grpo · llm · nlp · rl · scaling-laws

在 GitHub 上查看原仓库