
This model is a WIP. Currently, it was trained on a small dataset of one pattern design of "forbidden pants" however can be prompted well for different kinds of designs. It is quite overbaked and needs to be retrained, which is quite evident at higher LoRA strength values.
Therefore:
I recommend a LoRA strength of 0.4 - 0.7 maximum, leaning more towards the lower values specifically.
If you enjoy this LoRA, (and the big butts you'll generate with it) please consider Donating BUZZ, it would mean the world to me, even a small amount. I have very little resources, and so I rely on these kinds of services to train these kinds of models for us all to enjoy :) Thank you
描述:
9 images, 14 epochs, around 1512 steps, LR 1.0, prodigy
训练词语: floral patterned flare pants,patterned flare pants,bell-bottoms
名称: lora-000014.TA_trained (1).safetensors
大小 (KB): 149690
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success