Follow me on X/Twitter for exclusive content: https://x.com/Citron_Legacy 😁
Description
Trained on CivitAI Trainer
Trained on: 83 images
Num of Repeats: 4
Number of Epochs: 10
Training Batch Size: 4
Total Steps: 830
Network Dim: 32
Network Alpha: 32
Keep Tokens: 0
Shuffle caption: false
FAQ
Comments (2)
Love your models, ever since the CAT 1.5 checkpoints!
One minor request, would it be possible to name your LORAs with a bit more information? Vivid.safetensors mentions the style, but not the author or basemodel. Something like Vivid_PDXL_(CAT).safetensors would clearly explain who made it and which model it applies to.
Would also prevent collisions/duplicates if you make the same LORAs for different base models!
Thanks!
That's good feedback. Hmmmm I agree that my Lora naming is kind dumb lol
For a lot of my Pony Loras I name them P1, P2, P3, etc as in "P1" = "Pony v1" The loras on this model post are particularly bad lol
The Flat Color one is named uniquely though "FlatAnimeP1"
I think... if I name them something like VividCAT that might work? Maybe VividCitron because Citron will be very unique?
I found that with ComfyUI on XL if the Lora name says "_CAT_" then it actually gets interpreted as prompting for "cat" not just invoking the Lora (Its probably my ComfyUI setup is bad). As a result my naming conventions have been messy lol
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.