This LoRA is designed to generate a single, clearly defined afterimage depicting a handjob.

However, the success rate may decrease when the penis is too small to leave enough space for the afterimage, or when using an FT model.

Increasing the prompt strength, for example with "(after image, handjob:4)", may improve the results.

For v0.0.0, the LoRA was trained on a total of 79 images: 40 original training images plus 39 additional cropped images.