Agree; to the expand on your points, even trying ultra simple prompt like [black circle on a white background] and then running the output through CLIP to get i

Agree; to the expand on your points, even trying ultra simple prompt like [black circle on a white background] and then running the output through CLIP to get its text outputs. Then try to narrow the iterations between the two to be predictable and controlled by you. Obviously understanding the code and data the model is based on is likely best path to better understanding what is going on; for example, last time I checked, Stable Diffusion was limited to 77 tokens. Ultimately though, possible paths to understanding are like unlimited and just as creative as any other aspect of the topic.