AI glossary
What is LoRA?
Short answer
Low-Rank Adaptation. Fine-tunes by training small low-rank matrices while freezing the base weights.
For example
Instead of retraining all 7 billion weights of a model, LoRA might train a few million extra numbers, small enough to share as a file of a few megabytes.
Learn it in a free lesson
Related terms
Questions
What is LoRA in simple terms?
Low-Rank Adaptation. Fine-tunes by training small low-rank matrices while freezing the base weights.
What is an example of LoRA?
Instead of retraining all 7 billion weights of a model, LoRA might train a few million extra numbers, small enough to share as a file of a few megabytes.
Where can I learn about LoRA for free?
AI in Bed explains it in the free lesson "Fine-tuning on a budget: LoRA & quantisation" (Advanced level, 6 minutes), with no sign-up.