Login / Signup

Efficient Large Models Fine-tuning on Commodity Servers via Memory-balanced Pipeline Parallelism.

Yujie LiuZhiquan LaiWeijie LiuWei WangDongsheng Li
Published in: HPCC/DSS/SmartCity/DependSys (2023)
Keyphrases
  • markov chain
  • fine tuning
  • statistical models
  • statistical model
  • limited memory
  • cost effective
  • main memory
  • computational power
  • computing power
  • neural network
  • model selection
  • fine tuned