Never forget the bribery of the criminals serving on the Supreme Court.

https://www.theguardian.com/us-news/2023/aug/11/clarence-thomas-undeclared-vacations-gifts-rightwing-donors

It is a shit show from top to bottom.

https://jacobin.com/2022/06/chief-justice-john-roberts-supreme-court-extremism-campaign-finance-corruption

I won't touch the proprietary junk. Big tech "free" usually means street corner data whore. I have a dozen FOSS models running offline on my computer though. I also have text to image, text to speech, am working on speech to text, and probably my ironman suit after that.

These things can't be trusted though. It is just a next word statistical prediction system combined with a categorization system. There are ways to make an LLM trustworthy, but it involves offline databases and prompting for direct citations, these are different from Chat prompt structures.

I'm sure they will eventually try to force ID's because it would be profitable for criminal data theft ads stalkers. This is all about corrupt money and exploitation. Billionaires are worthless parasites that have no right to exist in a Democratic system. Fuck the US fascist oligarchy party.

Oobabooga is the main GUI used to interact with models.

https://github.com/oobabooga/text-generation-webui

FYI, you need to find checkpoint models. In the available chat models space, naming can be ambiguous for a few reasons I'm not going to ramble about here. The main source of models is Hugging Face. Start with this model (or get the censored version):

https://huggingface.co/TheBloke/llama2_7b_chat_uncensored-GGML

First, let's break down the title.

• This is a model based in Meta's Llama2.
• This is not "FOSS" in the GPL/MIT type of context. This model has a license that is quite broad in scope with the key point stipulating it can not be used commercially for apps that have more than 700 million users.
• Next, it was quantized by a popular user going by "The Bloke." I have no idea who this is IRL but I imagine this is a pseudonym or corporate alias given how much content is uploaded by this account on HF.
• This model is based on a 7 Billion parameter dataset, and is fine tuned for chat applications.
• This is uncensored meaning it will respond to most inputs as best it can. It can get NSFW, or talk about almost anything. In practice there are still some minor biases that are likely just over arching morality inherent to the datasets used, or it might be coded somewhere obscure.
• Last part of the title is that this is a GGML model. This means it can run on CPU or GPU or a split between the two.

As for options on the landing page or "model card"

• you need to get one of the older style models that have "q(numb)" as the quantization type. Do not get the ones that say "qK" as these won't work with the llama.cpp file you will get with Oobabooga.
• look at the guide at the bottom of the model card where it tells you how much ram you need for each quantization type. If you have a Nvidia GPU with the CUDA API, enabling GPU layers makes the model run faster, and with quite a bit less system memory from what is stated on the model card.

The 7B models are about like having a conversation with your average teenager. Asking technical questions yielded around 50% accuracy in my experience. A 13B model got around 80% accuracy. The 30B WizardLM is around 90-95%. I'm still working on trying to get a 70B running on my computer. A lot of the larger models require compiling tools from source. They won't work directly with Oobabooga.

Hey there Lionir. Thanks for the post. Can the Beehaw team please look into copying or getting the creator of this bot to work here? https://lemmy.world/u/PipedLinkBot@feddit.rocks

I think the person that created that bot is somehow connected to the piped.video project. I know the whole privacy consciousness thing isn't for everyone, but this bot's posts are quite popular elsewhere on Lemmy.

FYI, the main reason to use piped.video links is that it is setup as an alternative front end for YT that automatically routes all users through a bunch of VPNs to help mitigate Alphabet's privacy abuses and manipulation.

We really need to crush the parasitic billionaire problem that funds the right and leverages convenient idiots of the world by stoking their prejudice and hate. There will always be some innocent whipping boy to toss in a camp. Pitchfork politics is never about the victims it is about controlling the conversation and distraction. Like in the USA, the only reason for the stupidity is to prevent closing the loopholes that enable the corrupt oligarchy. No one can investigate and make reasonable laws when they are confronted by a constant barrage of absurd nonsense. It just needs to be so inflammatory that no one can dominate the conversation with the message I'm posting now to regain any sense of rational control. The USA has 10% of the laws and protections of any other western nation. This is why the parasites exist, what they are funding, and what is being exported world wide. Russia proved that power is all about leveraging the convenient idiots. Australia and Japan have solved this issue already. Anyone with tens of billions of dollars is a worthless subhuman criminal.

Rule number one of buying a new car: get the dealer to disconnect the modem.

Cars should be entirely open source by government regulation. All software should be public and the manufacturer should be required to host and maintain a public toolchain that can reproduce the software and any revisions made. All of this should also get mirrored by the library of Congress and made publicly available as a second source indefinitely. This is about ownership. Digital rights are never okay to reserve. If I do not own everything I am only renting from the real owner. Proprietary goods are theft of ownership. It really is that simple.

"Hi Karen , this is HR. You can now log anonymous complaints about IT, by logging into this external website with your company credentials. We provide this for your security because IT is able to monitor in network communication."

OP is aware that the term hackers had, and still has, a completely different meaning in the community it originated from. A real hacker is someone with the skill to modify a codebase to suit their needs. The corruption of the term by corporate media is an attempt to steal the fundamental human right to share one's digital work freely by labeling these people and their work as criminal. Hacking means modifying source code to suit your specific needs. It could be used to refer to the efforts of someone learning to become a kernel developer or it could be someone modifying the software source code in ways that are not suitable to share with others, such as features that may be incompatible with some systems or features.

Cracking has long been proposed by hackers as a more appropriate term describing a bad actor attempting to gain unauthorized access to a system.

It seems pedantic at first and in my simple explanation here too, but if you look into the details, philosophy, politics, and people where this term originated from, then look at the opposition and how they profited, and finally look at the state of current society and the internet, you will likely see this as, at least mildly, offensive. Like, I still misuse the term as if it has dual meanings, but anyone using it correctly gets my attention right away.

They need to hit the final nail on the head. All smart phones sold in Europe must have fully documented and open source hardware including the entire chipset, all peripherals, and the modem, with all registers and interfaces documented, the full API, and all programing documentation along with a public toolchain that can reproduce the software as shipped with the device and updated with any changes made to future iterations as soon as the updated software is made available.

This law would make these devices lifetime devices, if you choose; as in your lifetime. It would murder the disposable hardware culture, and it should happen now. Moore's law is dead. The race is over.

...officials are working to get a remotely operated vehicle that can reach a depth of 6,000 meters (about 20,000 feet) to the site as soon as possible.

The 5-person submersible, named Titan, is capable of diving 4,000 meters or 13,120 ft. “with a comfortable safety margin,” OceanGate said in its filing with the court.

but...after looking up on Wikipedia

...a wreck that lies over 12,000 feet (3,700 m) below the surface...

There are several vulnerabilities in bootloaders that have not been fixed. Namely, there is an entire tiny operating system that is used to initialize the processor before the main bootloader begins. Then the bootloader creates a bunch of handles to control the hardware, and hands them over to the operating system kernel. The i-core/ryzen processors include the second generation of this tiny underlying operating system that runs before everything else. This tiny operating system was originally marketed as a way to remotely monitor and troubleshoot data center servers, but this is a very weak and flawed marketing strategy. The way this system runs before everything else, it has root/admin privileges and access that supersede everything that comes after it. A bad actor accessing this system is absolutely game over for all hardware including the bootloader itself. Well this tiny operating system is tied to the microcode for the processor generation.

If you know anything about old computers that had a bunch of boards and chips inside the case, modern computers still have all of these chips and systems, but they are all integrated into just a few chips. These systems are still complicated and have a certain way they must be powered up and initialized so that each system begins in a specific state along a long chain. The "microcode" in a modern computer is really just a bunch of "software" that controls the order that the hardware is brought online. In the i-core/ryzen generation of hardware the microcode is proprietary and copyright protected. This is a way to get around many x86 patents expiring. It has long been speculated that the tiny operating system is also a back door for governments as it can completely own any system regardless of encryption or any other security measures.

There is a way to mostly disable this tiny operating system but there is no way to monitor or confirm its activity at run time. OpenSIL is like having access to the control room of this tiny operating system for the first time. It means it is now possible to completely secure and verify the state of a system. There is no security in obscurity. OpenSIL is the removal of a major failed attempt at security through obscurity.

Ultimately, at the most fundamental level, openSIL means full ownership over your hardware. I can buy AMD, but can only rent Intel. Intel keeps ownership of this tiny little corner of the hardware and they have done a terrible job of managing what they own. When faced with that buying choice the outcome should be obvious, assuming you are able to run the software that can take advantage of this. It will take at least a few months, but there should soon be a completely open source version of Coreboot that will use openSIL. Hope this helps.