AOM
7th layer
HeavenOrangeMix
I haven't tried it, but it probably doesn't work with Waife Diffusion π
There are too many to carry! Basically, the bigger the %, the bigger the impact π.
https://huggingface.co/Default38693/LoRA_AzurLane_Akashi
Q.What is the difference between these versions of Lion, AdaFactor and DAaptation...?
These are differences in learning, so LoRA users don't need to worry too much π.
If I had to say, Lion is probably the best.
Q.The uploaded LoRA hashes do not match!
Oops, sorry about that! I embedded a thumbnail and the hash has changed! Different hash, but same LoRA!π«
Q. What are the study settings?
.json is in π€! You don't want to go all the way to AN extension to look at it, do you?
Q.Where is the toml you used for training?
The base is in π€!
Q.I would like something other than the uploaded epoch!
Let's go π€
Q.What changed the learning rate?
Let's go π€
Q. It seems to contain tags that have a small impact and are unnecessary!
Her launch date is November 1897, so I may have unintentionally added it myselfπ
Q What Trim.safetensors!?
This is probably the safetensors version of the model you know!