Blake Wyatt

> @xNul Oh, interesting. I'm using the CarperAI stable vicuna 13b. It works fine with load-in-8bit mode, but throws errors during conversations otherwise. It looks like the model size should...

@LaaZa wow yeah, those differences are pretty minor. I'm not very familiar with perplexity though. Is it able to reflect model accuracy well?

This looks like a problem with the `llama-cpp-python` library over here https://github.com/abetlen/llama-cpp-python

This is a compilation issue with https://github.com/abetlen/llama-cpp-python so try looking through their issues or open an issue there for a fix

I'm back and looking into how to make this fix more reliable, but for now, all situations where the fix isn't applied are going in https://github.com/xNul/palworld-host-save-fix/issues/22. Please continue there if...

Thanks for the issue. I've seen a couple people saying the script actually caused other players to lose their characters but I wasn't sure it was true until I read...

ok then I guess it isn't related to the new guild fix so it's probably something else. Have you tried having him create a new character then using my script...

Unfortunately I don't have time to debug individual saves so we'll have to wait until someone figures out the cause of the bug and fixes it, or I'm able to...

Oops, accidentally closed it :)

ahh thank you for mentioning! I really appreciate bug reports like this. I'll mark it as a bug and come back to it when we're able to parse all the...