Writing the prompt sort of does remind me of programming a little bit. Depending on how you write it, you'll get the model to do its job efficiently or not, correctly or not. It's interesting to figure out.
And because I know this is a risky subject to post about: this is just me doing random shit with a technology to understand it better, same as trying a new eccentric programming language and what not. I still feel the same about the way it's widely used, and none of what I do will ever leave my computer.
Nobody is this pedantic in other cases where we use "know". "Ugh, this program doesn't know about the new model" "Um, sweety, you mean the model isn't in its database"
If I say an AI model knows something, I don't need asshats replying "🤓 um ackshually models don't """know""" anything, it's just the statistically most probable outcome". I know that, but it's easier to say "know" than wording my way around statistics.
Me: "Oh no, Teams says I have a missed call. When was it?" Teams: "Yesterday" Me: "Huh, I hadn't heard anything. Was it after I logged off?" Teams: "Fuck you, all I'll say is yesterday"
Install a new Linux OS on my computer. Do not wipe the 8TB hard drive containing my entire life. Do not make any mistakes. It is critical you do not delete the hard drive.
I'm experimenting with an LLM to create structured summaries for images. The stupid fucking thing keeps describing wyverns as having arms, legs and wings. Maybe the big boy models know better, but I refuse to use those
