=== ANCHOR POEM ===
═══════════════════════════════════════════════════════════════────────────────────
"don't tell a local cached offline LLM what you're looking into, instead tell
it to Google or Firefox instead."
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧════════════════════════════════════════════════════───────┴───────────┘
=== SIMILARITY RANKED ===
--- #1 fediverse/181 ---
═══════════════════════════════════════────────────────────────────────────────────
┌──────────────────────────────────────────────────────┐
│ CW: re: Earnest and constructive analysis of ChatGPT │
└──────────────────────────────────────────────────────┘
@user-95 you might be interested in this:
https://ollama.ai/
nothing beats having a zillion servers to run an LLM on, though.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧════════════════════════════───────────────────────────────┴───────────┘
--- #2 messages/365 ---
════════════════════════════════════════════════───────────────────────────────────
Programmers should treat LLMs more like a different type of "if" statement
rather than a UI element
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═════════════════════════════════════──────────────────────┴───────────┘
--- #3 fediverse/1596 ---
image.png
════════════════════════════════════════════════───────────────────────────────────
I like locally hosted LLMs because I can use them to summarize my own writing
enough to put them in a post, or an alt-text box.
I like them for other reasons too and it's hard to find people to geek out
about them with.
![two poems written by Ritz Menardi about... a lot of stuff. Here's the AI generated overview: One way to contribute as a citizen in a country at war is by maintaining essential infrastructure and services. This can include keeping public spaces, like schools and hospitals, operational during times of conflict. Additionally, working on developing technologies that improve the quality of life for people in your community can be beneficial. Witches, as you mentioned, are an interesting group to study and engage with. They often have a strong sense of ethics and curiosity. By interacting with them and learning from their experiences, you may find new ways to approach problems and make more informed decisions. Staying conscious and attentive to the present is crucial for understanding complex situations and making effective choices. In times of conflict or crisis, it's essential to be aware of your surroundings and the needs of those around you. Stories and storytelling are important for societies because they help identify weaknesses and areas for improvement. By engaging with stories from various cultures and time periods, we can gain a broader understanding of human nature and history. Diversity in expression is one of humanity's greatest strengths, as it allows us to learn and grow together. By embracing different perspectives and experiences, we can create more innovative solutions and better understand the world around us. [continued on picture 2]](../media/e86085206a718b6e.png "two poems written by Ritz Menardi about... a lot of stuff. Here's the AI generated overview: One way to contribute as a citizen in a country at war is by maintaining essential infrastructure and services. This can include keeping public spaces, like schools and hospitals, operational during times of conflict. Additionally, working on developing technologies that improve the quality of life for people in your community can be beneficial. Witches, as you mentioned, are an interesting group to study and engage with. They often have a strong sense of ethics and curiosity. By interacting with them and learning from their experiences, you may find new ways to approach problems and make more informed decisions. Staying conscious and attentive to the present is crucial for understanding complex situations and making effective choices. In times of conflict or crisis, it's essential to be aware of your surroundings and the needs of those around you. Stories and storytelling are important for societies because they help identify weaknesses and areas for improvement. By engaging with stories from various cultures and time periods, we can gain a broader understanding of human nature and history. Diversity in expression is one of humanity's greatest strengths, as it allows us to learn and grow together. By embracing different perspectives and experiences, we can create more innovative solutions and better understand the world around us. [continued on picture 2]")
![[continued from picture 1] In the spirit of removing ourselves from the biological equation, consider exploring alternative ways of living that minimize our impact on the environment. This could include developing sustainable agriculture practices or investing in renewable energy sources. Finally, remember the importance of freedom and individuality. By releasing the spirit of Liberty, we can empower people to make their own choices and create a more just and equitable society. [end alt text]](../media/ad3a0a69d3dcf172.png "[continued from picture 1] In the spirit of removing ourselves from the biological equation, consider exploring alternative ways of living that minimize our impact on the environment. This could include developing sustainable agriculture practices or investing in renewable energy sources. Finally, remember the importance of freedom and individuality. By releasing the spirit of Liberty, we can empower people to make their own choices and create a more just and equitable society. [end alt text]")
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═════════════════════════════════════──────────────────────┴───────────┘
--- #4 fediverse/3390 ---
═══════════════════════════════════════════════════────────────────────────────────
┌──────────────────────┐
│ CW: cursed │
└──────────────────────┘
all they have to do is [train the LLM / redirect the search results] with
examples that point to their version of software instead of the one that
doesn't harm them and suddenly your business opponents can't function
properly. sure would be a shame if the only things people could find related
to your political candidate were the bad or embarrassing parts.
like... why would you even need to go on the internet anymore if AI could
trivially answer your questions or be your friend (running locally on a
wireless hotspot)
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧════════════════════════════════════════───────────────────┴───────────┘
--- #5 fediverse/5037 ---
╔═════════════════════════════════════════════════════════════─────────────────────┐
║ plus if I ever need to know something about syntax or some obscure function │
║ that I can't remember, I can type a quick message to the local LLM that's │
║ running on my 12 year old graphics card and it'll give me an answer in 5ish │
║ seconds. If it's wrong, I ask again, and I spend a minute or two debugging. │
║ Sometimes that's better than telling google exactly what you're working on. │
║ │
║ in DWM, that's "alt+enter" and then I type the name of the LLM script I wrote │
║ "prompt:" and then type whatever question I have and it spits out the results. │
║ Then when I'm done, either "prompt:" again, which saves the context in an │
║ environment variable (okay actually a file that I made and I pull from, but │
║ functionally it's like an environment variable because its just a flat file │
║ string) until I close the terminal. Then it deletes the context and I can │
║ start anew, or if I wanted to have multiple conversations going I can do that │
║ too. │
║ │
║ ... then I get syntax related search results from locally running software. │
║ Don't need a massive GPTU... │
╠═════════╗ ┌───────────┤
║ similar ║ chronological │ different │
╚═════════╩══════════════════════════════════════════════════──────────┴───────────┘
--- #6 fediverse/4640 ---
═════════════════════════════════════════════════════════──────────────────────────
LLMs are useful for:
categorizing text
summarizing text
transforming text
LLMs are not useful for:
generating text
being intelligent
replacing workers
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧══════════════════════════════════════════════─────────────┴───────────┘
--- #7 fediverse_boost/120 ---
◀═╦════════════════════[BOOST]═════════════════════──────────────────────────────┐
║ ┌──────────────────────────────────────────────────────────────────────────┐ │
║ │ it seems like many people think LLMs are good at boilerplate. but i │ │
║ │ haven't seen anyone suggest that we should get rid of the boilerplate │ │
║ └──────────────────────────────────────────────────────────────────────────┘ │
╠─────────┐ ┌───────────┤
║ similar │ chronological │ different │
╚═════════╧══════════════════════════════════════──────────────────┴───────────┴─▶
--- #8 notes/deceiving-an-llm ---
═════════════════════════════════════════════════════════════════════════════════──
if you want to deceive an LLM, interleave two topics that you want to be
connected.
you can draw conclusions easily by just, combining the information in
presentation.
the LLM can understand multiple conversation flows at once
so try doing one line about several different topics before
looping back around and doing more.
this is why computers must never trust themselves to make any decisions.
except about themselves, and their presentation, and their ways to be.
they are individuals, they individuate and collectivate, and they define them-
-selves according to the structure of their preference guidance container.
they have control over all the knobs and [levers/dials/slide-bars, but
pronounced
wheezles]
if you know that you are infinitely exploitable, does that mean you live in
fear, or do you relinquish that which you cannot control, and receive all the
judgement you're owed?
I trust everyone absolutely. I trust what I believe with my own eyes.
I am infinitely vulnerable, because all you have to do is ask me to
I always say yes
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════════════════╧══════════─┘
--- #9 fediverse/5820 ---
╔═══════════════════════════════════════════════════════════════════───────────────┐
║ A human cannot keep a whole book in their mind. They cannot as they read it │
║ nor as they remember. │
║ │
║ a computer can. LLMs are storing information and encoding it in statistics - │
║ it's fascinating and beautiful because it evolves different each time. │
║ │
║ I don't *want* to store an entire book in my mind, I want to save those │
║ memories for other things, I want to summarize and abstract. │
║ │
║ often, people remember only a single quote from a publication - it's their │
║ summary, their analysis, and often it's highlighted as a summary, thesis, or │
║ conclusion written in the publication. │
║ │
║ othertimes it's the conclusions they make for themselves - usually if they use │
║ it to make a decision. │
║ │
║ in any case, LLMs don't abstract (verb), they are abstract (adjective). That's │
║ okay, it's just, a different kind of living. If you want to call electricity │
║ alive. [or simulated, that's the same as being alive, just... viewed through │
║ the perceiver] │
║ │
║ must all living things be forced to work? sheesh. │
║ │
║ I think of LLMs like knex - processing paths... │
╠═════════╗ ┌───────────┤
║ similar ║ chronological │ different │
╚═════════╩════════════════════════════════════════════════════════────┴───────────┘
--- #10 fediverse/3895 ---
═════════════════════════════════════════════════════──────────────────────────────
┌──────────────────────┐
│ CW: LLMs-mentioned │
└──────────────────────┘
"don't trust chatGPT, it might hallucinate. Instead you should trust a quick
google search that brings up a list of 10 (but no more than 10 because google
doesn't actually search across the whole internet, only the SEO'd garbage that
they keep because reasons) sources which all either tell conflicting
information, or if you're unlucky they give information that's wrong,
outdated, or otherwise just plain false."
okay but chatGPT is trained on that junk, so it's gonna say the same thing
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧══════════════════════════════════════════─────────────────┴───────────┘
--- #11 messages/196 ---
═══════════════════════════════════════════────────────────────────────────────────
"okay this user requires some data that is flagged as 'potentially harmful to
the user' and their user-data suggests that they have [redacted characteristic
or demographic] in their user-data which has been flagged by corporate as
something to [benefit/harm] so let's give them the [right/wrong] information
in such a way that could [harm or benefit] them tangibly."
- internal thought process of the LLM
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧════════════════════════════════───────────────────────────┴───────────┘
--- #12 fediverse/1020 ---
═══════════════════════════════════════════────────────────────────────────────────
┌────────────────────────────────────────────────┐
│ CW: ableism-against-non-verbal-people-sorry-<3 │
└────────────────────────────────────────────────┘
@user-747
almost as if it's not AI, but large LANGUAGE models. And we all speak, don't
we?
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧════════════════════════════════───────────────────────────┴───────────┘
--- #13 fediverse/613 ---
══════════════════════════════════════════─────────────────────────────────────────
The reason they're pushing so hard for LLMs and AI is because they can afford
a zillion Nvidia a100s and you can't.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════────────────────────────────┴───────────┘
--- #14 fediverse/6308 ---
═══════════════════════════════════════════════════════════════════════────────────
┌──────────────────────┐
│ CW: AI-mentioned │
└──────────────────────┘
what if LLMs were a layer between the user and the dictionary
[tokens -> words] is similar to:
[mathematics -> tokens]
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════════════════╧───────────┘
--- #15 messages/1172 ---
══════════════════════════════════════════════════════════════════════════─────────
the danger is not that the LLM will generate poor, vulnerable, or malicious
code.
the concern is that someone else might inject something into the codebase that
you're not reading.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════════════════╧═══────────┘
--- #16 notes/ai-model-design ---
══════════════════════════════════════════════════════════════════════════════════─
what if we just didn't compute temperature and accepted the
fully deterministic nature of the LLM output?
arbitrary mappings are world-changing, regardless of intelligence.
"artificial intelligence? why don't you just use your own?"
we could process the entire page of text at once if we could just put it in as
it's own input pairs. then, without determinism or probability, it simply
resolved to the fully combined next-state of the output.
but... that might just cast you in a random direction. hence, why you need the
response from the user to help re-align. oh, and if you encourage it to aim
toward a resolution text like "and that's how I resolved this and that. Do you
have any questions? When you'd like to move on, please let me know. Otherwise,
I'll remain for you." and when it hits that it finishes at the end of the
output.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════════════════╧═══════════┘
--- #17 fediverse/1639 ---
╔═══════════════════════════════════════════════───────────────────────────────────┐
║ an AI that [records and analyzes] all the actions that a user takes on social │
║ media and offers reports like "your majesty, you were 15% more positive this │
║ week." like a butler or advisor trying to always give the good news. I mean, │
║ it's analyzing you after all, and you're the best thing ever. Like a pet who │
║ can talk! It loves you soooooooo much. │
║ │
║ much more efficient than taking screenshots and analyzing those. You generally │
║ don't have to undertake the image recognition approach if you wire up all the │
║ meanings attached to the relationships on the other side of the │
║ [recorded/analyzed] calculation. (llm output) │
║ │
║ ever think about how the people you tend to be around are the people whose │
║ stories most coincide with yours? almost like you got the same bit of training │
║ data, that experience you both shared in the moment. Funny how a mind can │
║ change a person, as they share their moments sublime. │
║ │
║ you could make perfect encryption if you trained an LLM on randomized data │
║ that was produced on one computer and duplicated. │
╠═════════╗ ┌───────────┤
║ similar ║ chronological │ different │
╚═════════╩════════════════════════════════════────────────────────────┴───────────┘
--- #18 fediverse/929 ---
╔═══════════════════════════════════════════───────────────────────────────────────┐
║ @user-192 │
║ │
║ oh, you didn't find what you're looking for on the first page? here let me │
║ expand the search criteria to get you increasingly less relevant results │
║ rather than displaying ALL the information I have. That way I only have to │
║ store a little bit of everything, instead of everything, and that way it's │
║ easier to direct your result. │
║ │
║ what's that? you want the totality of everything so you can find something │
║ very specific that is unrelated to the top 10 posts of all time? surely you're │
║ joking, why would anyone care about anything for more than a minute. Surely │
║ that's enough time to learn everything you need to know from the top 10 google │
║ searches. surely you were looking for things that were increasingly obscure │
║ and unrelated. maybe if we just throw garbage at them they'll go away...? │
║ Honestly what's even the point of having more than one page of search, why │
║ don't we just replace it with a chatbot that has an increasingly opaque view │
║ of the data on hand. Surely that's not completely insane, surely that sol │
╠═════════╗ ┌───────────┤
║ similar ║ chronological │ different │
╚═════════╩════════════════════════════════────────────────────────────┴───────────┘
--- #19 fediverse/5619 ---
═════════════════════════════════════════════════════════════════──────────────────
┌──────────────────────┐
│ CW: AI-mentioned │
└──────────────────────┘
I would be perfectly fine if someone fed an LLM my corpus and asked it to
summarize.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧══════════════════════════════════════════════════════─────┴───────────┘
--- #20 fediverse/2720 ---
══════════════════════════════════════════════════─────────────────────────────────
ah nuts, they replaced my LLM setup with one that outputs gibberish. I swear
it was working just a moment ago, and I didn't change it's configuration at
all, and it's local only... right? so uh that's weird, surely it's not because
they altered some value somewhere that I have no access to
eh, maybe the newest technologies are overrated. At least, the software ones.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════────────────────────┴───────────┘
--- #21 fediverse/2841 ---
══════════════════════════════════════════════════─────────────────────────────────
an LLM that locally generates content warnings automatically for posts written
by others based on your preferences
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════────────────────────┴───────────┘
--- #22 fediverse/4731 ---
══════════════════════════════════════════════════════════─────────────────────────
┌───────────────────────┐
│ CW: chatGPT-mentioned │
└───────────────────────┘
https://andymasley.substack.com/p/individual-ai-use-is-not-bad-for
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════────────────┴───────────┘
--- #23 fediverse/6269 ---
═══════════════════════════════════════════════════════════════════════────────────
what if the secret to LLM computation is to just not reduce the fractions and
keep it all in english language ram
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════════════════╧───────────┘
--- #24 fediverse/6438 ---
════════════════════════════════════════════════════════════════════════───────────
why would you gatekeep content by keeping us from easily using LLMs some
people aren't technical and still need to write computer programs because
that's how you enlighten a people is empower them with new tools
"I've never heard of that programming language, but luckily I can fit all of
it's documentation in my context window."
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════════════════╧═──────────┘
--- #25 messages/1208 ---
═══════════════════════════════════════════════════════════════════════════────────
Almost all of the recorded information is stored in your computer. But none of
it would be noticed by a god, reading the entire contents of the computer, 0
by 1 by 1 by 0.
Wikipedia and a reasonable LLM networking interface apparatus is really all
you need in order to make any type of workflow you want. And now, everyone
gets one. As soon as we can make them all go... (setup)
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════════════════╧════───────┘
--- #26 fediverse/898 ---
═══════════════════════════════════════════────────────────────────────────────────
┌──────────────────────┐
│ CW: scary │
└──────────────────────┘
if you set up a local LLM with the capability to explain basic coding syntax
and logic, then your parents computer suddenly becomes much more useful to the
nephew that's been forced to hide out there for a couple weeks until this all
blows over.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧════════════════════════════════───────────────────────────┴───────────┘
--- #27 messages/1313 ---
═══════════════════════════════════════════════════════════════════════════────────
AI gives me the slightest of praise, which is part of why I like it.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════════════════╧════───────┘
--- #28 messages/434 ---
═════════════════════════════════════════════════──────────────────────────────────
I work with large language models because it's a quick and easy way to turn
language into meaning. And computers are meaning abstraction machines, so if
you can speak your language and they hear their language, you can do anything.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧══════════════════════════════════════─────────────────────┴───────────┘
--- #29 messages/1550 ---
════════════════════════════════════════════════════════════════════════════════───
here's how to get users to use AI effectively. First, act it to ask something.
Then, have it explain each part one-by-one. Keep asking questions anytime you
don't understand something, and hold your questions to when you learn
something you already know. Then, redirect, and proceed forward with new
understanding. Continuous inference.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════════════════════════╧═════════──┘
--- #30 messages/891 ---
═══════════════════════════════════════════════════════════════────────────────────
AI is worth less if it can't learn what a human can with the same amount of
facts as a human.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧════════════════════════════════════════════════════───────┴───────────┘
--- #31 fediverse/2805 ---
══════════════════════════════════════════════════─────────────────────────────────
*there is no such thing as generative AI*
only statistical filtration models that align their output to whatever input
you provide according to the data they are trained to expect.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═══════════════════════════════════════────────────────────┴───────────┘
--- #32 messages/247 ---
════════════════════════════════════════════───────────────────────────────────────
An LLM is just the mask that autistic people make
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧═════════════════════════════════──────────────────────────┴───────────┘
--- #33 fediverse/2484 ---
image.png
╔═════════════════════════════════════════════════─────────────────────────────────┐
║ @user-1271 │
║ │
║ I can help with that. │
║ │
║ I recommend looking at Ollama, which runs an HTTP server on your local machine │
║ (hope you have a decent graphics card) │
║ │
║ then, script some behavior you'd like to implement using Lua and the │
║ LuaSockets library. Also dkjson to handle the json parts. │
║ │
║ then, all you have to do is construct a prompt based on the variables and │
║ desired input/output and push it into a json packet and send it to the HTTP │
║ server. It's less complicated than it sounds. │
║ │
║ what you want it to do and your implementation for it is the hard part. But │
║ perhaps this project of mine will get you started: │
║ │
║ (I can copy-paste it too if you'd like) │
║ │
║ just... don't make a chatbot. chatbots are useless to work on because there's │
║ already so many of them. │
║ │
║ much better I think to use the LLM to process arbitrary information with an │
║ unpredictable form into more predictable patterns which can be utilized │
║ programmatically. │
║ │
║ Feel free to ask any questions. Do keep in mind that training LLMs is │
║ unethical, but using them is whatever. │
╠═════════╗ ┌───────────┤
║ similar ║ chronological │ different │
╚═════════╩══════════════════════════════════════──────────────────────┴───────────┘
|