No special tags needed, just prompt normally while the LoRA is enabled. "Hires fix" is recommended for best results.If you use this LoRA as part of a mix or host it on a generation service, please mention and link back to this page (especially if you're making money off of it)

Silverjow Style LoRA

<ul><li>470 images</li><li>Tagged with wd14-swinv2-v2</li><li>Tagged image repeats:<ul><li>33 1x repeat "monochrome, lineart"</li><li>17 1x repeat "normal quality"</li><li>8 1x repeat "sketch"</li><li>101 2x repeats "high quality"</li><li>311 2x repeats "masterpiece, best quality"</li></ul></li><li>x768 base resolution with aspect ratio bucketing</li><li>25 epochs, batch size 5</li><li>1e-3 cosine with restarts LR (matching U-Net and text encoder)</li><li>25 cycles</li><li>256 dim</li><li>Clip level 2</li><li>[redacted] base model</li><li>[redacted] .vae enabled during training</li></ul>

Files with SHA256: a05301d1805edf262e83e1b569a68c004e1bb7616fea56ba68de37dd4599d4ec