Login / Signup

ModuLoRA: Finetuning 3-Bit LLMs on Consumer GPUs by Integrating with Modular Quantizers.

Junjie YinJiahao DongYingheng WangChristopher De SaVolodymyr Kuleshov
Published in: CoRR (2023)
Keyphrases
  • image coding
  • general purpose
  • parallel processing
  • coding scheme
  • real time
  • data sets
  • neural network
  • vector quantization
  • vector quantizer
  • quantization scheme
  • entropy constrained