SD 2.0 embedding for animalselfie trained text inversion with default settings and 20 images for 1000 steps. captions were things like: animalselfie of a dog at the beach

example prompts: