I made this model using Seaart AI's Lora training function. I have no idea how this really came out compared to the one's I've used Google collab to make, since the Lora training function does not have much information on how to configurate the more advanced settings. The only tutorials I found is the one they have on the Seaart AI subreddit and the other's are in another language. If you can add Trigger words with Seaart I don't know how.
For this model I suggest using lavender skin & Gloria Sato in your prompts for this model. Euler a seems to be the best sampler since I used Cartunafied to train them. Lora weights used best around 0.8-1. Mess around with it and let me know what works best.