where is the code to actually fine tune with qlora?
#4
by
brando
- opened
where is the code to actually fine tune with qlora?
This model was trained with the finetune_guanaco_65b script found in the QLoRA repo I linked to in the model card. That script is essentially just an argument list for qlora.py which is the script that performs the actual finetuning. So that is the script you should look at if you want to train your own model using the repo. It has a list of all the available arguments and what they do near the top.