A Dreambooth model trained on top of YiffAI 2.2.20 with 203 images of Cthugha from Tokyo Afterschool Summoners. Images were captioned with e621 tags without underscores or commas.<h3>Positive Tags</h3>Captions all began with:<ol><li>safe/questionable/explicit</li><li>cthugha (tas)</li><li>dragon</li></ol>Most captions included:<ul><li>anthro</li><li>male</li><li>solo</li><li>red body</li></ul>Other tags included in captions:<ul><li>headshot portrait</li><li>bust portrait</li><li>half-length portrait</li><li>three-quarter portrait</li><li>full-length portrait</li><li>low-angle view</li><li>high-angle view</li><li>side view</li><li>rear view</li><li>fully clothed</li><li>topless</li><li>bottomless</li><li>nude</li><li>penis</li><li>tentacles</li><li>torn clothing</li></ul>The outfit in his first skin is tagged with "cthughajacket cropped jacket". His second is tagged with "cthughacoat coat".<h3>Negative Tags</h3>It seems to have trouble with his eyes. More than anything I use empty eyes in the negative prompt.

Files with SHA256: c5c13d78d905576a6a4e262c49b566204049f7173255f229cf2157b351d3a35e