Dense Training, Sparse Inference: Rethinking Training of Mixture-of-Experts Language Models
arxiv(2024)
摘要
Mixture-of-Experts (MoE) language models can reduce computational costs by
2-4× compared to dense models without sacrificing performance, making
them more efficient in computation-bounded scenarios. However, MoE models
generally require 2-4× times more parameters to achieve comparable
performance to a dense model, which incurs larger GPU memory requirements and
makes MoE models less efficient in I/O-bounded scenarios like autoregressive
generation. In this work, we propose a hybrid dense training and sparse
inference framework for MoE models (DS-MoE) which achieves strong computation
and parameter efficiency by employing dense computation across all experts
during training and sparse computation during inference. Our experiments on
training LLMs demonstrate that our DS-MoE models are more parameter-efficient
than standard sparse MoEs and are on par with dense models in terms of total
parameter size and performance while being computationally cheaper (activating
30-40
DS-MoE-6B model runs up to 1.86× faster than similar dense models like
Mistral-7B, and between 1.50× and 1.71× faster than comparable
MoEs, such as DeepSeekMoE-16B and Qwen1.5-MoE-A2.7B.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要