This upload is for testing Hugging Face workflows, runpod and vastai templates soundness, and how character cards I've written test out. Not a branch for actual model. Uploaded as part of a personal LoRA and infrastructure prototyping process.

Please see argilla/CapybaraHermes-2.5-Mistral-7B for it's Repo details.

Model notes: Brilliant little thing. Hilarious. Stubborn. Not very steerable. I've had some hilarious, hairpulling with mild exasperation arguments with it: "Cappy! Stop trying to be a goblin and focus, damn it! OOC!" (literal argument I've had laugh). But very open. Just has opinions and ways of doing things. Love it. Quite helpful once I can get it's attention to focus on what I am saying, not whatever smart-arse roll it's on. Just like working with humans! Takes on RP with it's own flair/interpretation, which I also love. Brings a lot of personality to roles.

File:

License: OpenRAIL++