大模型训练技术论文_gpu efficient large-scale language model training

A Reading List for MLSys

An Overview of Distributed Methods | Papers With Code

ZeRO: Memory Optimizations Toward Training Trillion Parameter Models

https://arxiv.org/abs/1910.02054

Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM

https://arxiv.org/abs/2104.04473

Reducing Activation Recomputation in Large Transformer Models

https://arxiv.org/abs/2205.05198

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

https://arxiv.org/abs/1909.08053

Fully Sharded Data Parallel: faster AI training with fewer GPUs

Fully Sharded Data Parallel: faster AI training with fewer GPUs Engineering at Meta -

GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding

https://arxiv.org/pdf/2006.16668.pdf

GSPMD: General and Scalable Parallelization for ML Computation Graphs

https://arxiv.org/pdf/2105.04663.pdf

Automatic Cross-Replica Sharding of Weight Update in Data-Parallel Training

https://arxiv.org/abs/2004.13336v1