Heard some more and more people are trying out local LLM's in their rigs. I've started exploring that avenue late 2023, ...

Heard some more and more people are trying out local LLM's in their rigs. I've started exploring that avenue late 2023, back when llama.cpp and other forks were transitioning from ggml to gguf's for CPU+GPU based inference and I got four words: "Be safe" and "stay sane"It's all I can say, lol. It'd help if they have friends to ground them back to reality.I'd be afraid if they have none. The things these do to your brain when you're alone. If you're easily impressed and you can't separate fiction from reality, you will definitely fall in the trap of "praising what the "AI" can do" until you wish everything gets replaced by "AI", that every hurdle can be solved by "AI". No.. it's not that simple. It's easy to think that way cos life's hard but that's not it.You'll know you're in too deep if you've thought about resorting to offloading the ability to think to an LLM and wish others to do the same as well...#mastodon #ai #llm #localai #localllm