Description:

A sketchbook ink-and-wash look. Lines become loose, visible pencil and ink strokes, colour goes on like light watercolour over textured paper, and details get a hand-drawn, storybook feel while the picture stays clean and readable.

This isn't a trained LoRA. No images were used to make it. It comes from steering neurons inside the base model:

• I mapped the UNet's feed-forward neurons into groups. This file is three groups of about 1,000 neurons each, across three layers (one early in the UNet, two style layers), all pushed down, mixed by a random combination search.

• The LoRA stores that push as a bias change, which is why the file is only about 17 KB.

How to use

• No trigger word. Just load the LoRA; it acts on the model only, not the text encoder.

• Strength 1 is the look in the previews. The effect scales linearly with strength: lower it for a lighter touch; negative values push the other way.

• It stacks with other LoRAs.

• Preview settings: 768×1024, 20 steps, CFG 4.5, DPM++ 2M Karras.

Compatibility

• Made and tested in ComfyUI (Load LoRA / LoraLoaderModelOnly) on prefectIllustriousXL v4.0.

• Other UIs: the file holds only bias changes, which some UIs (A1111 / Forge) may not load. If nothing changes, that's why.

• Other Illustrious checkpoints: untested, but they share the base, so it may carry over.