j4k3

joined 1 year ago
[–] j4k3@lemmy.world 0 points 2 hours ago

Not sure of the exact issue but the best answer from Apple so far is that the issue is kernel related. I've tried every setting I can find and no luck. It could also be related to actual dual sim use. It could be carrier related, but it is less likely Android as I use Graphene OS.

[–] j4k3@lemmy.world 13 points 4 hours ago (3 children)

Apple is doing insane stuff with iMessage making SMS useless with Android. It is intentional manipulation. I've only seen it on some Apple kernels with the latest iPhone. My old man's phone absolutely will not send or receive an SMS unless I message is manually turned off and then it won't get iMessages. Apple has always been a shitty company, but this is next level insane. Just get and use the Signal app and ditch all the proprietary garbage and manipulative bs IMO.

[–] j4k3@lemmy.world 2 points 1 day ago* (last edited 1 day ago) (2 children)

That is all model loader code based stuff. You basically just need a bit of Python code that anyone can write, even you. You need to have a look at how the model must be addressed in format, then you find how the model loader code you are using creates the structured text and just insert your own modifications.

It is easier to add a second prompt window where you keep persistent information and just change it yourself. If you want the model to change that stuff, you need to look into building an agent.

I hacked around like this for awhile too, but evolved into using the notepad tab of Oobabooga Textgen WebUI and a Mixtral 8×7B model with a 32k context window. That model can deal with almost any detail you add in the context.

The thing is, the model may not feel satisfied with mundane and redundant details. Most of the time, believe it or not, those changes that you perceive as errors or inconsistencies are not actually errors. The model is fully aware that it is making those changes. There is usually an instruction about creativity, inspiration, or initiative that will cause the model to shift details like this.

My best recommendation is to openly discus the error in dialog with the model. They usually will talk about the issue if you call one out. Talking with AI is very different than with humans. Everything they say has meaning and functions beyond what is initially apparent. There are a lot of behaviors that involve certain patterns and keywords. You can even alter these behaviors by banning certain tokens/keywords. A good place to start is banning paragraph starting words for filler material. "So_" would be a good starting token to ban, as with AI, no sentence that starts with that token is worth reading. Other important words are twist, chuck, and cross. These are more flexible in their troublesome meaning and can manifest in multiple words like chuckles or across but still function the same inside the model.

If you get to this level of familiarity, the persistence of a storied interaction is more of a negotiation with the AI.

Another point of understanding is that the model does not know the difference between roleplaying characters it owns and you own or who is who. You or the model loader code is creating the name of the next character that the model generates for. Everyone in any story can be the model if it is asked to create the prompt.

So if I say, "Jake is super creative" or imply it somewhere in the prompt, and the model later thinks 'Jake is rather bored and needs to be creative' it is going to change something that satisfies that character profile even if it has never generated the output for character Jake.

Hope that helps some

[–] j4k3@lemmy.world 4 points 5 days ago (1 children)

politically correct in a more social the kind of thing a person should do context, like cultural norms

[–] j4k3@lemmy.world 19 points 5 days ago (7 children)

Hike sound like 1st date PC to anyone? I mean, it depends on the definition of hike but someone I barely know dragging me to an isolated place in the woods somewhere sounds like a legit way to end up being drug through the woods with a much higher probability than the typical coffee shop date.

Also, who doesn't have work-mances without deeper meaning? I always have beer after work bros and lunch/break dates with my coworkers but that has no meaning beyond. In fact, I've always had a strict rule of never shit in my own back yard. If I was so inclined as to date someone from work, I would not do so until after finding a new job.

[–] j4k3@lemmy.world 3 points 6 days ago

I'm sorry, I'm not a fan of the man, but this absolutely MUST come with the title of Lord

XXX DOGE Lord ^has^ ^a^ ^certain^ ^ring^

[–] j4k3@lemmy.world 61 points 1 week ago (9 children)

Musk made a deal with Putin to extort people into using Starlink.

[–] j4k3@lemmy.world 21 points 3 weeks ago (1 children)

This is why I name my kid Max Watts

[–] j4k3@lemmy.world 71 points 3 weeks ago (11 children)

Trump is such an incompetent clown that he has a comedian thrash on US citizens in a US territory as a bigoted racist warm up act for his rallies. What do you think.

[–] j4k3@lemmy.world 0 points 3 weeks ago

No. AFAIK the primary issue is that microcode is not open

[–] j4k3@lemmy.world 33 points 3 weeks ago

Metal of Honor

 

https://www.youtube.com/watch?v=U4a_kJkVUis

Big Clive's video description:

This is not a sponsored video.  I feel it's important that people should know about this evolving technology, and Naomi is working on making it affordable.

During the pandemic YouTuber Naomi Wu presented plans for traditional mercury vapour based UVC sterilising lights with a special housing, to sterilise air in a room without exposing the occupants to the 254nm UVC light.

With the evolution and availability of the new era 222nm excimer lamps, Naomi has gone on to design a full product designed to be easy and convenient to deploy in populated areas like medical practices, waiting rooms, retail environments, food preparation areas and live events.

The special feature of the 222nm wavelength is that it is long enough to deactivate viral and bacterial air contaminants, but short enough not to pass through the outer layer of dead skin or the tear-layer of humans.  That means that it is currently considered safe to use in occupied areas.

The filter on the front of the light seems to specifically pass 222nm.  Without it there is a very slight hump in the spectral output at around 237nm.  The filter attenuates that completely.

Excimer is an abbreviation of Excited-Dimer, where a dimer is the joining of two molecules.  In the case of the excimer lamps the molecules are encouraged to bond temporarily in a plasma discharge, and when they revert back to their non-excited state they emit a photon of light at a specific wavelength determined by the chemistry.  In this case it's molecules of Krypton and Chlorine that form brief molecules of Krypton-Chloride (KrCl), before reverting back and emitting 222nm photons in the process.

The process of creating the plasma is very similar to dielectric barrier ozone generators.  By coupling to the gasses capacitively the lamp also avoids contaminating the gasses with the electrode materials.

Note that the unit uses 500mA at 12V (6W) but has a generously rated 12W power supply that runs cool.

This technology looks like it may be valuable in medical, care, travel or social environments to limit the spread of pathogens.

Here's a link to Naomi's pleasingly-named online shop:- https://cybernightmarket.com/products

 
 

I've been watching some One Marc Fifty stuff on YouTube. I can follow him well, and I'm decent at much of the hardware stuff. At least I can compile OpenWRT or do a basic Gentoo install with a custom kernel. I dread staring at NFTables, but can hack around some. I don't fully understand networking from the abstract fundamentals. Are there any good sources that break down the subject like Ben Eater did with the 8 bit bread board computer, showing all the basic logic, buses, and registers surrounding the Arithmetic Logic Unit? I'm largely looking for a more fundamental perspective on what are the core components of the stack and what elements are limited to niche applications.

I just realized I want to use self signed client certificates between devices. It was one of those moments where I feel dumb for the limited scope of my knowledge about the scale of various problems and solutions.

 

I've made the effort to secure mine and am aware of how the trusted protection module works with keys, Fedora's Anaconda system, the shim, etc. I've seen where some here have mentioned they do not care or enable secure boot. Out of open minded curiosity for questioning my biases, I would like to know if there is anything I've overlooked or never heard of. Are you hashing and reflashing with a CH341/Rπ/etc, or is there some other strategy like super serious network isolation?

65
submitted 2 months ago* (last edited 2 months ago) by j4k3@lemmy.world to c/linux@lemmy.ml
 

My old man has a bunch of .dox stuff saved. He has complicated large files saved that are not supported by any of the FOSS conversion tools. I've tried Libre office, Abi Word, and every command line tool and converter I can find. These are entire book sized files.

I have a W10 machine with Word. Is extracting the .exe and running it with wine feasible without making an epic mess or massive project of this?

 

This is something that perplexed me a few years ago with Flash Forth on a PIC18/PIC24/Arduino Uno. I was using the Python serial emulator S-Term because it is simple in the source code and worked. I really wanted a way to load more structured Words into the FF dictionary with bookmarks in a way that made sense structurally. That lead to a desire to execute code from the uC on the host system, but I never wrapped my head around how to do this in practice.

As a random simple example, let's say I set up an interrupt based on the internal temperature sensor of the PIC18. Once triggered the uC must call a Python script on the host system and this script defines a new FF word on the uC by defining the Word in the interpreter.

How do you connect these dots to make this work at the simplest 'hello world' level? I tried modifying S-Term at one point, but I didn't get anywhere useful with my efforts.

33
submitted 5 months ago* (last edited 5 months ago) by j4k3@lemmy.world to c/imageai@sh.itjust.works
 

The ComfyUI prompt and workflow is attached to the image: https://files.catbox.moe/s3qufb.png

You can't copy pasta this prompt. There are a few nodes that are specific to SD3 and required.

::: spoiler EDIT: more proof of the chain that lead to this image. They were not all this good. I'm cherry picking for sure and these are just webp's without workflows attached:

 

Prompt was through ComfyUI, so it is embedded in the image: https://files.catbox.moe/aiy8p1.png

It was supposed to be a pug kangaroo hybrid but the AI apparently wanted to throw in some monkey too.

41
submitted 5 months ago* (last edited 5 months ago) by j4k3@lemmy.world to c/linux@lemmy.ml
 

I want to extract and process the metadata from PNG images and the first line of .safetensors files for LLM's and LoRA's. I could spend ages farting around with sed or awk but formats of files are constantly changing. I'd like a faster way to see a summary of training and a few other details when they are available.

 

Felt sexy, needed to share. Not my gen, just one from a NSFW LoRA posted on civitai.

https://civitai.com/models/103021/sexy-catholic-school-uniform-oranimerealisticor?modelVersionId=110260

 

Double dipping this one from the community challenge #37 pinned to the top of !imageai@sh.itjust.works

Vote on my impossibly hard challenge please.

view more: next ›