The LoRa was trained using in-game screenshots of Yennefer. At higher weights, she appears more like the game character, while weights below 0.8 allow for more natural and realistic images of her to be created.
The LoRa was trained on just two single transformer blocks with a rank of 32, which allows for achieving such a small file size without any loss of quality. Since the LoRa is applied to only two blocks, it is less prone to bleeding effects.