Verified: a year ago
SafeTensor
Type | |
Stats | 205 |
Reviews | (17) |
Published | Feb 26, 2024 |
Base Model | |
Training | Steps: 1,600 Epochs: 16 |
Usage Tips | Clip Skip: 1 Strength: 0.9 |
Trigger Words | 0xellenrose |
Hash | AutoV2 310B5D2BB9 |
Created on Civitai
Ellen Rose, born 1988, is an British YouTube content creator. The LoRA was trained on 50 publicly available photos in 1024x1024, 896x1152, 832x1216, and 768x1344.
Recommendations
Use trigger 0xellenrose and LoRA weight 0.8 - 1.0
Photorealistic models such as Juggernaut XL tend to produce the best results.
Trained with Clip Skip 1, results without Clip Skip vary.
Positive prompt dyed hair may be interesting.
I didn't do much testing with CFG values, but 7 to 8 seems to be fine.