Sign in Publish

What is a LoRA config?

What is it?

LoRA adapts an existing model to a task without fully retraining it: only small adapters are trained. A LoRA page gathers all the settings of that training.

When to use it

  • Reproduce a fine-tuning that worked.
  • Start your own training from proven settings.

Page fields

MethodLoRA, or QLoRA (4-bit compressed model, less memory).
Base modelThe model to adapt, e.g. Qwen/Qwen2.5-7B-Instruct.
Rank, alpha, dropoutSize and strength of the adapters.
Learning rate, epochs, batchTraining settings.
Linked datasetThe catalog dataset used for training.

How to use it

“Use” tab: download the YAML config for Axolotl, or copy the PEFT Python example.

Browse LoRA →