torridgristle

Yeah, that sounds right. Either feeding it no optimized latent code at all in some layers, or lerping it back to the original latent code to some extent in some...

Sigma for 1/2 results meeting the old range: LI: 0.042 RE: 0.085 NO: 0.170 CA: 0.255 EX: 0.425

I'm stoked to see it, not even just for using it with this project but just to figure out how to do this sort of task at all. Please tell...

Here's how to fix it: Edit the script to add `, *args` to the all the *_get_learned_conditioning commands, like `def hijacked_get_learned_conditioning(model, prompts, steps, *args):` on line 7 and `return real_get_learned_conditioning(model,...

Appears to work if I disable "Move VAE and CLIP to RAM when training if possible. Saves VRAM." after getting this message. Previously I had used the "Unload SD checkpoint...

This might be a shot in the dark, but if you still have the embeddings could you try using the negative embedding in the positive and the positive in the...