Positive weights give a photorealistic style, and negative weights give a cartoon style.
Trained using weights from -5 to 5.
Description
Initial model trained using https://github.com/ostris/ai-toolkit
FAQ
Comments (2)
When you say, "trained using weights from -5 to +5". What does that mean in terms of training? How do I caption images to achieve this "slider" effect?
Thanks! nice lora btw
This LoRA was actually trained without any images, only text! It uses the slider training code from here: https://github.com/ostris/ai-toolkit