PixelPilgrim
@PixelPilgrim@lemmings.world
- Comment on Fediverse Corporate Sabotage 4 weeks ago:
More like we will need a game plan for them. Best I can say is that admins should be open about whatever legal troubles they get into and we should have redundancy in as many countries as possible.
They will fight so we should expect, plan, prepare
- Comment on The YouTube Alternative Nobody's Talking About ! Peertube 5 weeks ago:
just peertube recommendation algorithm browser extension. Working pretty well and ik discovering channels that make/reupload well produced content
- Comment on Peertube video recommendation program 1 month ago:
of cource i forget to log in to this account after posting. but so far Im building it to be more like a browser extension that collects your watch history and video meta data peertube video via api and locally suggests what video to watch next
- Comment on Peertube video recommendation program 1 month ago:
of okay. i guess if i get to the point where aggregating data from multiple users then yeah ill look into publishing the data somehow, most of the data will be pulled from peertube instances
- Comment on Peertube video recommendation program 1 month ago:
im thinking of something more like a browser extension that modifies webpage with recommendations, so far its 1 person (myself) then build up from there. i dont think peertube instance would have to be involved outside of just giving the meta data for local videos
- Comment on Peertube video recommendation program 1 month ago:
yeah i hope people build on my work and improve it since i doubt ill even have this project working anytime soon.
i have hard time concieving of p2p system for this. is it like heres what this person likes so do “X”
- Submitted 1 month ago to fediverse@lemmy.world | 11 comments
- Comment on The best thing *you* can do for the fediverse is *just be kind* 1 month ago:
Yeah the merit of a stance is different than someone’s ability to argue. They’re different concepts.
- Comment on The best thing *you* can do for the fediverse is *just be kind* 1 month ago:
So all your saying is don’t engage with bots (even though it’s fine because bots can get you information quickly), and that you can’t ignore negative people online.
- Comment on The best thing *you* can do for the fediverse is *just be kind* 1 month ago:
Lol so you agree with this post I made just not the other
- Comment on The best thing *you* can do for the fediverse is *just be kind* 1 month ago:
Lol my argument automatically has no merit without you hearing any of it. That’s like bigotry or something.
I guess Ben Shapiro argues with people that aren’t media trained to make his stance look better, do you want me to say people Ben Shapiro argues make some awesome argument all the time without exception?
I’ll teach you that it’s not on me to make your arguments, get your evidence. I just put it on myself to communicate my arguments and poke holes in other people’s arguments
- Comment on The best thing *you* can do for the fediverse is *just be kind* 1 month ago:
Ikr, I like pissing people off then making much of them for not being able to defend their position. I think it’s good for engagement
- Comment on The best thing *you* can do for the fediverse is *just be kind* 1 month ago:
I generally say bullet points are good ideals, but there’s a much bigger issue with mental health.
There are certain people in lemmy that need to learn what “you have no enemies” and “I’m gonna do my own thing” means. It’s fine people are different live on earth is very diverse
- Comment on The basics of the Fediverse. Facebook, Twitter, Tiktok, Reddit, ... vs the Fediverse. What do you think? 2 months ago:
It’s actually a graphic. I’d put a legend explaining some of the fediverse icons. And 2 sentences explaining that each fediverse icon communicated with other fediverse icons
- Comment on Have you said Thank You once? 2 months ago:
The AI models are so nice I swear at them when an issue I cause by not reading the the instructions they have and so is just like “I know your getting frustrated, let’s try this approach” and calmly repeats the instructions.
- Comment on There is a live stream of fedi plays Dragon Warrior Monsters 2 months ago:
So you not like dalek.zone instance?
- Comment on Is 33 cents a small amount of money? 2 months ago:
lol i wonder how much that is just guessing. they just coin flipped it
- Submitted 2 months ago to fediverse@lemmy.world | 4 comments
- Comment on playing around with desktop ai 2 months ago:
I’ll have to see if my setup can handle 14b. I havent found eBay Gutenberg but I found llama 3 guttenberg.
I’ll have to check out exllama, but I only have 1 gig of vram
- Comment on Reddit will warn users who repeatedly upvote banned content 2 months ago:
Seems like lemmyverse needs to advertise on Reddit
- Comment on Reddit will warn users who repeatedly upvote banned content 2 months ago:
So if the content is banned how are people up voting it?
- Comment on playing around with desktop ai 2 months ago:
I get annoyed when it makes post that rephrases the original post, but I think it helps that it’s generic then it’s less offensive and helps it blend in. Keeping it short helps
I’m thinking adding a program that puts in random grammar errors and an AI that does a LLM ai check to filter responses
- Submitted 2 months ago to fediverse@lemmy.world | 5 comments
- Comment on Preliminary LLM tests 2 months ago:
Cool then I will continue my development
- Comment on Preliminary LLM tests 2 months ago:
It’s interesting that the deepseek AI can blend in. It made a post like it really visited a waterfall. It didn’t say anything too specific as to make it obvious it wasn’t there(maybe that the weather was good in this time of year in March). So it will require looking at multiple posts or maybe even interacting to spot the bot.
These boys can definitely blend in and then run script that can push an ideology. I won’t make more than 1 since I don’t have the resources
- Comment on Preliminary LLM tests 2 months ago:
You can’t know any of that since you most likely haven’t interacted with the bot. Plus I don’t know “real option” is supposed to mean
- Comment on Preliminary LLM tests 2 months ago:
Rn I’m supervising the response and iterating the prompt. I haven’t even implemented a scraper or api. I haven’t fully botted it yet because I don’t know the architecture I want. Like do I even want ai detector or a script that puts in grammar mistakes or an basic algorithm to pick prompts
- Comment on Preliminary LLM tests 2 months ago:
I’ll give you some advice. It’s better to state why you don’t want me to test so at all. Asking questions just comes off as standoffish. Plus the way you phrases your question exposes that you have no intention of understanding me
- Submitted 2 months ago to fediverse@lemmy.world | 13 comments
- Comment on Is there any counter AI bots in the fediverse 2 months ago:
So far I’ve experimented with ollama3.2 (I don’t have enough ram for 3.3). Deepseek r1 7b( discovered that it’s verbose and asks a lot of questions) and I’ll try phi4 later. I could use the chat-gpt models since I have tokens. Ironically I’m thinking about making a genetic algorithm of prompt templates and a confidence check. It’s oddly meta