--- [🍷 Shelved in Worlds Bar]---
This LoRa is trained on concepts of semantic relations and order, and its dataset included a large number of visual technical pretension , UML and nodes visuals. thats why it acts like an affecting force more then a compositional element.
imagine this this as hidden force that need shows indirectly by affecting elements,this means it needs an array of the same element to work with, example water drops, spider web, dust particle, ui elements . . . .
its effect includes : ordering , rearranging as a pattern , forming relations links
you must use the trigger word NodesTech
to activate it .
a simple direct way to use it is for a virtual reality aesthetic, you do this by adding elements like data-space ,holographic
you can try adding floating nodes
to oush it more into abstraction
works rly well with energy and fluids LoRAs
healthy LoRA weight range from 0.8 to 1.5
send to img2img to flesh out more details or to strengthen the LoRA even more
I recommend checkpoints that works great with generating small details like "revAnimated"