A lora based on the character Cerberus from Last Origin

Trained with 167 images

Image info

• Model used anything4.5

• bad hands embedding

Notes

• Use the provided trigger words to get her outfits

• No trigger word is needed if you dont want any of her outfits

• the wedding dress outfit is slightly overbaked due to the low amount of images so lower the lora weight to 0.8 for it. (the rest of the outfits can be left at 1)

Training settings?

The most important part of training is getting a good dataset!

Training settings can be found by installing the additional networks extension in auto1111

Extensions -> available -> load from -> additional networks

Restart webui

place lora into stable-diffusion-webui\extensions\sd-webui-additional-networks\models\lora

To view the training settings

Go to additional network tab -> choose model -> training paramters