p4ss3r2

/

wan22_Pupil_Transition_LoRa

like 1

Thanks for sharing the LoRA weight. The results are really impressive!
I was wondering if it would be possible to share some details about the training hyperparameters you used. If it's convenient, it would be even more helpful to have the full training script (starting with accelerate launch ...) so that I can better understand and reproduce your setup.
Thank you again for your amazing work!