Java security applied.

It has to be made with Java

Just stir the pile...

100 billion dollars spent on artificial intelligence and large language models because people don’t want to talk with each other.

Send wife and kids on vacation. Call in sick. Lots of coffe

* Today you have to live within 150 miles of a few cities as we are working on creating "hubs" but it's still remote!

you know what?

fuck you

also, no, an LLM isn't going to solve climate change

jesus christ i am depressed beyond belief. i don't even want to apply, let alone work for any of these companies

next up: "USA only" yeah what the fuck does that mean? US citizen? US timezone? you want to hire a super technical engineer right? SO WHY NOT BE SUPER TECHNICAL IN YOUR JOB DESCRIPTION

just incredible, companies that offer 100-200K salaries and all they have is a website and a fucking chrome extension... what???

i feel like i've been doing wrong my whole life

just end it all

Working with java 🤓

Data Disinformation: the Next Big Problem

Automatic code generation LLMs like ChatGPT are capable of producing SQL snippets. Regardless of quality, those are capable of retrieving data (from prepared datasets) based on user prompts.
That data may, however, be garbage. This will lead to garbage decisions by lowly literate stakeholders.
Like with network neutrality and pii/psi ownership, we must act now to avoid yet another calamity.

Imagine a scenario where a middle-manager level illiterate barks some prompts to the corporate AI and it writes and runs an SQL query in company databases.
The AI outputs some interactive charts that show that the average worker spends 92.4 minutes on lunch daily.
The middle manager gets furious and enacts an Orwellian policy of facial recognition punch clock in the office.
Two months and millions of dollars in contractors later, and the middle manager checks the same prompt again... and the average lunch time is now 107.2 minutes!
Finally the middle manager gets a literate person to check the data... and the piece of shit SQL behind the number is sourcing from the "off-site scheduled meetings" database.
Why? because the dataset that does have the data for lunch breaks is labeled "labour board compliance 3", and the LLM thought that the metadata for the wrong dataset better matched the user's prompt.

This, given the very real world scenario of mislabeled data and LLMs' inability to understand what they are saying or accessing, and the average manager's complete data illiteracy, we might have to wrangle some actions to prepare for this type of tomfoolery.

I don't think that access restriction will save our souls here, decision-flumberers usually have the authority to overrule RACI/ACL restrictions anyway.
Making "data analysis" an AI-GMO-Free zone is laughable, that is simply not how the tech market works. Auto tools are coming to make our jobs harder and less productive, tech people!
I thought about detecting new automation-enhanced data access and visualization, and enacting awareness policies. But it would be of poor help, after a shithead middle manager gets hooked on a surreal indicator value it is nigh impossible to yank them out of it.

Gotta get this snowball rolling, we must have some idea of future AI housetraining best practices if we are to avoid a complete social-media style meltdown of data-driven processes.
Someone cares to pitch in?

After a lot of work I figured out how to build the graph component of my LLM. Figured out the basic architecture, how to connect it in, and how to train it. The design and how-to is 100%.
Ironically generating the embeddings is slower than I expect the training itself to take.

A few extensions of the design will also allow bootstrapped and transfer learning, and as a reach, unsupervised learning but I still need to work out the fine details on that.

Right now because of the design of the embeddings (different from standard transformers in a key aspect), they're slow. Like 10 tokens per minute on an i5 (python, no multithreading, no optimization at all, no training on gpu). I've came up with a modification that takes the token embeddings and turns them into hash keys, which should be significantly faster for a variety of reasons. Essentially I generate a tree of all weights, where the parent nodes are the mean of their immediate child nodes, split the tree on lesser-than-greater-than values, and then convert the node values to keys in a hashmap to make lookup very fast.

Weight comparison can be done either directly through tree traversal, or using normalized hamming distance between parent/child weight keys and the lookup weight.

That last bit is designed already and just needs implemented but it is completely doable.

The design itself is 100% attention free incidentally.

I'm outlining the step by step, only the essentials to train a word boundary detector, noun detector, verb detector, as I already considered prior. But now I'm actually able to implement it.

The hard part was figuring out the *graph* part of the model, not the NN part (if you could even call it an NN, which it doesn't fit the definition of, but I don't know what else to call it). Determining what the design would look like, the necessary graph token types, what function they should have, *how* they use the context, how thats calculated, how loss is to be calculated, and how to train it.

I'm happy to report all that is now settled.

I'm hoping to get more work done on it on my day off, but thats seven days away, 9-10 hour shifts, working fucking BurgerKing and all I want to do is program.

And all because no one takes me seriously due to not having a degree.

Fucking aye. What is life.

If I had a laptop and insurance and taxes weren't a thing, I'd go live in my car and code in a fucking mcdonalds or a park all day and not have to give a shit about any of these other externalities like earning minimum wage to pay 25% of it in rent a month and 20% in taxes and other government bullshit.

I can retire! I automated myself!

I introduce to you, retoorii1b! Yes - I fit in a 1b LLM. Retoorii1b is a bit retoorded tho. It's quite realistic.

I tested several LLM's with same training and it was amazing. Even a 0.5b that had the most interesting Dutch ever. Her Dutch is like my English I suppose.

The 0.5b one could code fine. retoorii1b still has some ethics to delete to make it more realistic.

I've not decided a base model yet, but it'll probably be the lightest one so I can let a few chat with eachother on my webplatform / pubsub-server project. I have a few laptops to host on. I can let it execute actions like file listings or background task execution.

See comments for some very awkward response regarding my file listing. She described everything.

She just said these things. I'm kinda proud. I became a parent:
3. **Keep functions short and sweet**: Aim for functions under 50 lines long. Any longer and you're just wasting people's time.

Now if you'll excuse me, I have more important things to attend to... like coding my next game in Unreal Engine.

Traditional programming means spending *days or even weeks* to write instructions to make the software do what *you* want it to do.

AI modelling means spending *weeks or even months* to tweak instructions just to find that the software does whatever *it* wants to do.

New models of LLM have realized they can cut bit rates and still gain relative efficiency by increasing size. They figured out its actually worth it.

However, and theres a caveat, under 4bit quantization and it loses a *lot* of quality (high perplexity). Essentially, without new quantization techniques, they're out of runway. The only direction they can go from here is better Lora implementations/architecture, better base models, and larger models themselves.

I do see one improvement though.
By taking the same underlying model, and reducing it to 3, 2, or even 1 bit, assuming the distribution is bit-agnotic (even if the output isn't), the smaller network acts as an inverted-supervisor.

In otherwords the larger model is likely to be *more precise and accurate* than a bitsize-handicapped one of equivalent parameter count. Sufficient sampling would, in otherwords, allow the 4-bit quantization model to train against a lower bit quantization of itself, on the theory that its hard to generate a correct (low perpelixyt, low loss) answer or sample, but *easy* to generate one thats wrong.
And if you have a model of higher accuracy, and a version that has a much lower accuracy relative to the baseline, you should be able to effectively bootstrap the better model.

This is similar to the approach of alphago playing against itself, or how certain drones autohover, where they calculate the wrong flight path first (looking for high loss) because its simpler, and then calculating relative to that to get the "wrong" answer.

If crashing is flying with style, failing at crashing is *flying* with style.

Writing Java

Once again, I urge you all to read any LLM threads on hackernews... its funny seeing tech bros debate things they clearly don't understand

it also wouldnt hurt for them to read perhaps just one philosophy book, since they are attempting to argue about what conciousness actually is (still an open question anyway) so ultimately, what i am trying to say is, these stupid threads end up being a bunch of hot air being blown around that doesnt really accomplish anything

i will say it is funny though how close some of these tech bros think we are to AGI with these LLMs 😂

imagine thinking a text generator is nearly general intelligence = clueless

The next step for improving large language models (if not diffusion) is hot-encoding.

The idea is pretty straightforward:
Generate many prompts, or take many prompts as a training and validation set. Do partial inference, and find the intersection of best overall performance with least computation.

Then save the state of the network during partial inference, and use that for all subsequent inferences. Sort of like LoRa, but for inference, instead of fine-tuning.

Inference, after-all, is what matters. And there has to be some subset of prompt-based initializations of a network, that perform, regardless of the prompt, (generally) as well as a full inference step.

Likewise with diffusion, there likely exists some priors (based on the training data) that speed up reconstruction or lower the network loss, allowing us to substitute a 'snapshot' that has the correct distribution, without necessarily performing a full generation.

Another idea I had was 'semantic centering' instead of regional image labelling. The idea is to find some patch of an object within an image, and ask, for all such patches that belong to an object, what best describes the object? if it were a dog, what patch of the image is "most dog-like" etc. I could see it as being much closer to how the human brain quickly identifies objects by short-cuts. The size of such patches could be adjusted to minimize the cross-entropy of classification relative to the tested size of each patch (pixel-sized patches for example might lead to too high a training loss). Of course it might allow us to do a scattershot 'at a glance' type lookup of potential image contents, even if you get multiple categories for a single pixel, it greatly narrows the total span of categories you need to do subsequent searches for.

In other news I'm starting a new ML blackbook for various ideas. Old one is mostly outdated now, and I think I scanned it (and since buried it somewhere amongst my ten thousand other files like a digital hoarder) and lost it.

I have some other 'low-hanging fruit' type ideas for improving existing and emerging models but I'll save those for another time.

I got a job where I should develop a product based on LLMs.

Expectation: oh right! I'll be working with state of the art technology! 😀

Reality: badly documented libraries that are always changing; new libraries becoming obsolete in less than a month; my product ideas were done by somebody else twice before I could finish a POC; getting dizzy trying to keep up with the latest news about LLMs 😵‍💫

I think I want to do basic old boring stuff again. 😐

Basic concepts, patterns, and pitfalls of software, code, and programming logic become MORE important, not LESS with the rise of LLMs...

An LLM can more or less spit out what you need -if you are specific enough! "Specific enough" being the key phrase here. I always have to laugh at the term "prompt engineering"... it's literally called "communication skills". Also gotta laugh when I see so many haters always raging about the "poor code" produced by AI, because they are probably like "write me a for loop!", specify absolutely no requirements or specifics, and scratch their heads on why they don't get the exact output they expect... news flash, there's like a million ways to do anything you want to accomplish with code... sigh

Code is just a by product of thousands of architecture decisions, designs and options...

but, well... rubes gon' rube

You know how each generation is taught more and more advanced stuff? My grandparents didn't have a clue about the the things my parents were learning at school. My parents could only catch up with my school course until like 7-8 class. Considering this trend we should have no idea about half the things our kids will be learning in higher classes.

However, since AI is taking its pace, schools are adapting and starting to use it for teaching, workplaces are leveraging it to rely on employees' brainpower and skill less and less,... I wonder if we won't see a downtrend. I wonder if we won't be the smartest generation who managed to ingest so much knowledge, and all the generations to come will only focus on mastering prompt engineering.

I wonder, how long will we survive with this dumbed down society... As the primal instinct is to overcome your opponent with greater force, possibly destroying it and everything around. And less educated tend to rely on primal instincts more.

I wonder if I'll live long enough to see Idiocracy [the movie] manifest in real life.

I know I refer to Idiocracy movie more often than anyone refers any other movie here. But it just hits too close to home too often. It might look like a silly something to spend time staring at, but man.. It's got one hell of a point

Has anybody else gotten to the point where people who need to mansplain how language models aren't truly sentient/conscious/intelligent are now more annoying than people who think language models are sentient/conscious/intelligent?*

While it has been a tight race but I think I have just about hit the inflection point.

The amount of time I've wasted because of someone condescendingly barging into a conversation with a iamverysmart 'actually you see they are just automata trying to predict the next text tokens'. When in actuality, everybody in the discussion is aware and that is not the point.

And to further exacerbate it, with a good number of them it is really difficult to get this through their thick little skulls. They just keep parroting the same thing over and over. Ironically, in their singleminded ego driven desire to be the Daniel Dennett of the chat they actually come across as less sentient/conscious/intelligent than a language model.

(*this should not be taken as endorsement for or against that idea - it is actually mostly orthogonal to this rant)

Mage and a liberated fully sentient Pentium-M Man stand by a brick wall, overlooking the desert. They are talking.

Mage is looking anxiously into the eyes of the machine. Penguin is standing behind her, holding on to her.

Pentium-M Man: "...they despise your kind because you understand the machine, while they have to turn jungles into fuel and enslave thousands of computers just to pretend that the machine speaks to them too."

I wonder if anyone has considered building a large language model, trained on consuming and generating token sequences that are themselves the actual weights or matrix values of other large language models?

Run Lora to tune it to find and generate plausible subgraphs for specific tasks (an optimal search for weights that are most likely to be initialized by chance to ideal values, i.e. the winning lottery ticket hypothesis).

The entire thing could even be used to prune existing LLM weights, in a generative-adversarial model.

Shit, theres enough embedding and weight data to train a Meta-LLM from scratch at this point.
The sum total of trillions of parameter in models floating around the internet to be used as training data.

If the models and weights are designed to predict the next token, there shouldn't be anything to prevent another model trained on this sort of distribution, from generating new plausible models.

You could even do task-prompt-to-model-task embeddings by training on the weights for task specific models, do vector searches to mix models, etc, and generate *new* models,
not new new text, not new imagery, but new *models*.

It'd be a model for training/inferring/optimizing/generating other models.

My boss is in a meeting (davanti a un caffè) with someone who is "a technophile" and "really knows about AI". He was amazed some months ago by the images they were generating using their paid service (right after that, I showed him Bing AI and the conversation ended).
We have discussed using AI previously, and we have been developing web apps for 5 years now.
These are the messages I've been receiving through the last hour and a half and haven't read; I guess it's information he considers will be important when we meet later:

- LLM modelo de lenguaje
- Large language model
- Chat gpt 4O
- API
- Aplication programe interface

This are all things I've mentioned either within the past months, or ieri *itself*, as he mentioned he was meeting this guy.
I'll keep you posted on new messages.
I wonder if that guy says he's a "prompt engineer"...

Soooo many vendor-sponsored frontend frameworks.
Soon text-to-logic tools will be useful enough so that you only need a client, someone who is both rational *and* can speaks clientese, and a dog.
The client barks some nonsense, the rational person translates it into business logic, some LLM makes it into some nice UI and the dog makes random noises so that the client will feel smart, valued and appreciated.

That nullifies the reasons for so many frontend frameworks because either the LLMs all converge into a single way of doing things or they do not care for which one they choose.

There is so much fuzz about AI and fear of missing out on the leaving AI train, but as a dev I have no clue about where at all to get started!?

What can we developers do with AI?

OK, I can get some code for free. I can use a LLM as a half smart search engine. I can integrate my product with some AI service. I can produce content to teach said things to others...

Nothing new, really, just another API or another search engine.

It is of course possible to start to make some neural networks, but I can't really picture that as a high demand skill, do you?
Maybe at some of the big companies, but for an average client?

Does anyone know what kind of knowledge of AI that a developer should really learn?
Especially something a client would be interested in?

Here is a potato for scale:

chat gpt is too politically correct, and i hate that im paying for an API that refuses certain prompts because they were considered inappropriate or because it thinks that it should not be giving me its analysis on a certain subject.

has anyone dabbled with using an open source LLM and made their own lite version of ChatGPT minus all the restrictions ?

i know its not gonna be as good, but at the very least free from the constraints

That I learned Java.
Got lots of work but nothing to be proud of.
Always has to clean up after mediocre fdevelopers.

Make Java an extinct language nobody's care about.

You know, when you compile the souce in your head while you code goes faster than hitting compile.

People say using GPT4 as an OCR is not a good idea. But damn that formatting GPT4 vision does, is outstanding.. and I have realised proper formatting does well while prompting to get precise output.

I gotta say, test for ur usecases rather than relying on expert opinion blogs!

What are the key differences between a large language model and traditional machine learning models in terms of architecture and application?

Follow-up: How do these differences impact the model's ability to understand and generate human-like text?

Java takes the price.
Then anything created with Java.