Type | |
Stats | 731 |
Reviews | (66) |
Published | Jan 8, 2024 |
Base Model | |
Training | Steps: 135 Epochs: 405 |
Usage Tips | Clip Skip: 1 |
Trigger Words | abb3yl33 |
Hash | AutoV2 81EC2C95AE |
A commission of @hmonk.
Abbey Lee is an Australian actress and model, known for her roles in films such as Mad Max: Fury Road and TV shows like Lovecraft Country.
135-step TI trained on a dataset of 15 images with these settings. For this particular TI I've increased the token count to 8 (from my usual 4), hence the slightly higher than usual weight of the embedding. I'm liking the results and haven't noticed so far a decrease in the TI's flexibility (which does happen frequently with higher token count TIs), so this might become the norm from now on.
Curious about my work process? I have summarized it here.
Do you have a specific idea for a TI in mind? Visit my website and let me know.
Building a good prompt with my TIs
You're obviously free to experiment, but bear in mind that my TIs are trained with a more or less fixed phrasing, that normally starts with:
"photo of EMBEDDING_NAME, a woman"
So I recommend always starting your prompt like that and then building the rest of the prompt from there. For instance, "photo of beautiful (abb3yl33:0.99), a woman, perfect hair, upper body, (modern photo, graphic tee), 24mm, (analog, cinematic, film grain:1.3), (at a red carpet event), ((detailed eyes)), (color picture:1.1), (looking at viewer)".
Please also note that I'm using the "add detail" LoRA for my example pics. I recommend setting it around 0.5 for best results. CLIP Skip should always be set at 1 when generating pics with this TI.