Shaymin Land Forme LoRA trained on NovelAI with Kohya training scripts using the DAdaptAdam optimizer at 512x512 resolution. Works pretty well, but adding (large bushy hair, hair flower) to my prompt improved consistency. If your model tries to add a tail, put (tail) in your negative and (tailless) in your positive. Should have wide compatibility with model mixes based on NovelAI, which is most of them. Trained on an all female dataset.