Since my last embed was getting really mixed results with many people, I decided to retrain with different settings.
PLEASE USE WITH THE "vae-ft-mse-840000-ema-pruned" VAE!!!
and DO NOT OVERLOAD with negative prompts, sometimes they affect the embed and you will LOSE likeness.
This Millie Bobby Brown embed was trained in SD 1.5, for 200 Steps using Textual Inversion at a learning rate of 0.5:50,0.05:100,0.005:200
Batch size of 10, Gradient Acc of 9 which is an effective 18K Steps. The Dataset contains 90 images. Embed is 3 vector.
Type the name of the embed in prompt to use in your gens, its "MilBob3vLR" by default but you can rename it to whatever you please.
I'd really appreciate it if you could take the time to leave a rating/comment after you try out the embed! Helps me a lot and also tells me if I should be tweaking or changing anything for future uploads or versions!
Thank you in advance <3
I also have a Ko-fi for tips and commissions