💬Quantized Low-Rank Adaptation (QLoRA)
Also known as QLoRA, Quantized LoRA
LoRA on a squeezed 4-bit copy of the model. Now it fits on one card.
Think of it like
It is like packing a big coat in a small bag. Then you sew a patch on it.
Example
Emma wants to tune a large open model, but she has one 48 GB card. With QLoRA the squeezed model fits, and only the small add-ons are trained.
Why it matters
Now people can tune big models at home, with no data centre.
Definition
Common mistake
Key takeaways
- 1Squeeze the base. Frozen weights in 4-bit.
- 2Train LoRA on top. The same small add-ons.
- 3One card. Big models fit in less memory.