Show newer

@garbados My knowledge is a bit higher than that, maybe at the level of some of the types of chemicals and physical layouts. Is this related to the solar panels still or an altetnative to needing oil?

@skryking If those are irreplaceable by other materials than a society without access to oil would be pretty screwed 😅

@garbados very little. Just the basics of how batteries work for powering basic circuits.

I wonder if solar panels could be feasibly built by a society that had no access to oil and internal combustion. Seems like a useful skip if possible.

It's even more fun when the instance out is me and then I have to decide when in my schedule I'm gonna fit in fixing postgres. 🤪

Show thread

Wild seeing instance outages in real time from this side of the bridge

My glasses chain broke and I feel so naked without it.

For anyone curious, here's the published example app. Set up ollama and Agregore and you're good to go!

hyper://agregore.mauve.moe/docs/examples/llm-lenses-chat/

Show thread

I'm getting a lot of mileage out of the "strange and esoteric" filter.

Might just make it a permanent addition. Can't be too much slower than the average ad blocker free page load.

Original thread: mastodon.mauve.moe/web/@freaka

Show thread

Facebook and misinformation. 

@CyberpunkLibrarian Social media is so funny cus like, they somehow managed to trick users into writing all the content for them, now they want us to be the mods? What's next, we're supposed to host the site too? Actually, come think of it, that's just Mastodon...

@freakazoid Yeah that's kinda what I do, I have the model output some specific text then I turn it into something machine readable. e.g. this code for generating the html/css/js for a page: github.com/RangerMauve/llm-app

Getting output data out of an LLM outside of raw chat interfaces is like scooping water with a leaky bucket. You need to plug more and more holes and add extra buckets to catch the spills. Structured JSON can help for a lot of cases but it makes the text generation less capable.

Fun example in the app I made for it. The message the LLM actually interpreted can be seen when you mouse over. I should find a name for a simulation of the experience of being misunderstood.

Show thread

@blaaablaaaa @feliks One could set the tonal lens to something like "Angry and rude" and then every post you see would be rephrased to be angry even if they weren't originally

@feliks Whoops, sorry I forgot to link the original, edited the post

@feliks Yeah sure, here's an example of a donald trump quote through the lens "kind and caring"

Original text: x.com/realDonaldTrump/status/1

Just made a tool that uses local in @agregore to rewrite the "tone" of text on a page with a user defined filter. You can now either up the stress level of your doom scrolling or apply rose tinted glasses with a single click.

LLM "alignment" is so annoying. They basically shut down and shut up whenever they encounter negative text. I'm trying to make a filter that rephrases hate comments as complements but it just refuses to engage if the negativity is high enough 🤷

Being a 31337 h4x3r can be handy for bypassing web base "not allowed to run on linux" checks.

New: Meta has deleted trans and nonbinary Messenger themes, as well as the blog posts announcing them. Happens the same week that it has changed its rules to allow users to say LGBTQ+ people are "mentally ill"

404media.co/meta-deletes-trans

Show older
Mauvestodon

Escape ship from centralized social media run by Mauve.