Training
QLoRA: Efficient Finetuning of Quantized LLMs
University of Washington·May 23, 2023
Tim Dettmers, Artidoro Pagnoni, Ari Holtzman
View on arXivTL;DR
Combines 4-bit quantization with LoRA to fine-tune large models on a single consumer GPU.
Why it matters
Made fine-tuning big models accessible to almost anyone, supercharging the open community.