Here's the video where I learned about this: <a target="_blank" rel="ugc" href="https://www.youtube.com/watch?v=0ChoeLHZ48M&amp;ab_channel=LatentVision">Higher quality images by prompting individual UNet blocks</a>This research is being on the model: <a target="_blank" rel="ugc" href="https://civarchive.com/models/405630/zavyfantasiaxl-pdxl">ZavyFantasiaXL - PDXL - v2.0</a><hr />Sometimes we want more control over the concepts/styles we're trying to prompt, but the bleeding is too high on some concepts or models based on pony. Sometimes the solution is to use different start/end steps for those concepts, but that's not always enough.I was looking for a solution and found an interesting one. It works by putting a concept in different blocks of the model. The trick is to find which blocks are in charge of your concept.This technique can be a bit difficult, not perfect, and might become obsolete in the near future, but I want to share my research with you. I don't know the technical details behind this, but maybe someone can enlighten us. Without further ado, let's get to the samples.<hr />On the left is the normal prompt, and on the right is the prompt injection.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/8d2ed7b0-8ace-48ea-a59a-b370e66d2bdb/width=525/8d2ed7b0-8ace-48ea-a59a-b370e66d2bdb.jpeg" />Positive: score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, score_safe, solo, upper body, (chibi:1.9), (human:1.8), black hair, black eyes, beautiful eyeline, biting_own_lip, squinted eyes, big breasts, holding breasts, red blouse, dnw, castle backgroundNegative: logo, print, furry, gloves, nsfw, nipples, score_6, score_5, score_4<hr />Now, I'll list what I found that each block affects in the composition of my image. I need to create more images with different compositions and landscapes for a more comprehensive conclusion, but I hope this helps someone and that you share your own findings too.<ul><li>IN4, IN7, IN8, MID0, OUT0, OUT1, OUT3, OUT4: score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, solo</li></ul>I add my quality tags to all the blocks that affect significant elements.<ul><li>IN4, IN7, MID0, OUT3, OUT4. Update: I'm doing more research on these blocks, because it looks like they affect the overall style somehow, but not like OUT0.</li><li><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/120e6947-5f54-4845-b70e-cd3dda1cce10/width=525/120e6947-5f54-4845-b70e-cd3dda1cce10.jpeg" />OUT3, OUT4: It looks like these blocks affect the ambient occlusion or details slightly, because disconnecting them results in this:<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f53f9ea2-b8de-44d9-8b5f-d5ac7092ca2b/width=525/f53f9ea2-b8de-44d9-8b5f-d5ac7092ca2b.jpeg" /><s>Sometimes prompting in these blocks doesn't make strong changes, and other times it affects your image a lot, so I usually leave these empty. If I need a little change, I add a few random word</s></li></ul><ul><li>IN5, OUT2, OUT5: biting_own_lip, squinted eyes</li></ul>I not sure which one of these blocks affects the face, but when I add the concept "biting..." it works better than just using the tag on other blocks.<ul><li>BLOCK IN8: chibi, upper body, blouse, castle</li></ul>This block takes a brute concept without details, only the things you want to be in your composition.<ul><li>BLOCK OUT0: score_safe, (chibi:1.9), big breasts, upper body, biting_own_lip, holding breasts</li></ul>This block affects the body, pose, hairstyle.<ul><li>BLOCK OUT1: (human:1.8), black eyes, biting_own_lip, red blouse, nipples through clothes, beautiful eyeline, upper body, dnw, black hair, castle background</li></ul>Here you can be more descriptive. This takes the concepts you add on IN8 and adds the color and style. Since this block has a lot of weight on the composition, I set a start step of 0.10.

Files with SHA256: 66707dd69c3c283522bfd0218ca8c0f23bb91ae03cbae33ac4cea4ee3259f0b8