Use the trigger 'final boss'. Animation values between -0.8 and 0.8
Trained on:
"final boss++:0.4|final boss%{random_prompt}:-0.1"
For 2000 steps using https://github.com/ntc-ai/conceptmod
"The stories we tell ourselves shape who we are. We become the heroes, the villains, and the final bosses in the tales of our lives. Choose your story wisely, for you may find yourself playing a role you never intended." - gpt4 misquoting Patrick Rothfuss
(technical) Tutorial https://civarchive.com/models/50070/mania-concept-trained-without-data
Please react to my videos to build rlhf dataset: https://civarchive.com/user/ntc/images?sort=Most+Reactions
Description
FAQ
Comments (4)
Interesting! Can it only be used for animation or can it still make stills?
Training results in the lora and the animation is the lora being applied between -0.8 to 0.8. I will edit the description to link to the tutorial but it's difficult to setup right now.
That one friendly NPC at that start of the game turns out the be the Final Boss.
if you use this lora at 2.0 strength your stable diffusion will become skynet