This is a "from below" slider LoRA, and it's the first model where I've put my custom layer-editing tool to real use. The tool lets me fine-tune individual layers of a LoRA, and for this one I stripped everything down to just 2 of the 30 layers a zImage LoRA contains. That single change is what made this slider possible. Without it, I don't think this LoRA could have existed at the quality it has now.
You don't need to prompt "from below" for it to work, but adding that kind of phrasing will reinforce the effect. When paired with character LoRAs or other heavy modifiers, you'll likely need to push the strength higher to get the angle to come through.
The model breaks down quickly once you cross its limits, and where those limits sit depends on your prompt and what else you have loaded.
In a clean prompt with nothing else fighting it, the sweet spot is around 6 to 8.
If your prompt already leans into a "from below" angle, that same 7 to 8 range will take it extremely low.
Stacked with heavy LoRAs, you may need to push it up to 9 or 10 before it starts hitting the limit.
For best results use a prompt that leans into "from below" and use around 5-6
Sidenote:
I think im going to add the range in the filename from now on, since my loras often have very high range compared to others. So this lora is named from-below_max8_loraholic
Please give me feedback if this is something you guys would like me to continue to do.