An exemple to fine-tune a Mistral 7B model using Quantisation, PEFT and LORA
The training data is booth grosly insuficient and not the best choice as the model is already working fine on that subject.
It is just an exemple to demonstrate the process
kolergy/Train_Mistral_7b
An exemple to fine-tune a Mistral 7B model using PEFT and QLORA
Jupyter NotebookMIT