果冻甜甜的

12-28

Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism

12-28

Reducing Activation Recomputation in Large Transformer Models

09-05

pytorch中的stream和event

08-24

attention中张量并行与GQA

08-24

pytorch send and recv

08-24

pytorch Shard

08-24

pytorch中TCPStore Rendezvous机制

08-24

ubuntu常见shell命令

08-24

ubuntu搭建技术博客指南

08-24

lumos:Efficient Performance Modeling and Estimation for Large-scale LLM Training