Hi, Thanks for the feedback. I'm sorry for the inconvenience; I'm still learning how this works. I trained this LoRA in Tensor.art. I don't know what that hash is. Let me know if I can change it, and how to do it.

I have another Sandman model, but it's even worse because it was trained in LyCORIS, so it's not working here... But I guess it's OK to download.

Here are the training parameters, just in case they help:

Seed: -

Clip Skip: 1

Text Encoder Learning Rate: 0.00001

Unet Learning Rate: 0.0001

Learning Rate Scheduler: constant

Optimizer: AdamW8bit

Network Dimension: 64

Network Alpha: 32

Gradient Accumulation Steps: -

Tag Parameters:

Mix Description: -

Save N Tokens: -

Advanced Parameters:

Noise Offset: 0.03

Multires Noise Discount: 0.1

Multires Noise Iterations: 10

conv_dim: -

conv_alpha: -

Batch Size: -

Sample Image Settings:

Prompt: photography of SandmanNM, a tall, gaunt, bony humanoid with pale, almost white skin. He has hollow black holes instead of eyes and no mouth. The figure has black forearms, black hands, and long, claw-like fingers, and wears a black long skirt and silver medieval knight armor with crossed arms. Outdoors, eroded broken castle ruins and debris on the ground, sunset, golden beams lighting.

Sampler: euler