adinhitlore

Sorry, I replied. Spacex failed with Mars.

in theory several GBs .bin may be useful chatbot if it's not transformers even better. Not too easy to code but...for me as a user it's preferable over SaaS right now.

qwen, grok, chat ji pi ti? They are close to sonnet in quality...if you want quality drop, no high standards: mistral or solar 3 pro or something.

This is also an interesting case against LLMs/chatbots. If it was 1903 chatbots would've repeated the same that airplanes are impossible or that time is constant, there is no attraction between diamagnetic materials (casimir effect) or anything experimentally proven later. I'm not saying one should pursue claims that may always remain pseudoscience such as magnet motors but it helps to always doubt common academic consensus at the time and rather: experiment.

* fastest human-made object (well, for the time being it's failing);

* this too: https://github.com/bggb7781-collab/lrnnsmdds

* SNN, NLP spiking neural network;

* neurosymbollic code generator (half-abandoned, not quite feeling like resurecting 2010s and amazon alexa is the right choice atm).

she looks gemini generated?? either musk made the profile "so cool, legalize comedy" or false flag from altman? 20% chance for a legit retarded doomer.

it just clicks with me, i like the syntax of the language or i believe it may solve important problems (e.q. AI helping with cancer cures, helping engineers design propulsion, improve loneliness etc). Existing documentation was important too back in the days before 2023 and the AI boom, for example asking question about c# on stackoverflow is one thing but asking about the Ring programming language totally different. Finally: versatility of language, i tend to avoid limited purpose languages like say perl or ruby which may be amazing idk but lack of gui tools, ai ecosystem, gaming development, app development etc i'd rather avoid them.

i hope he does, nothing but a puppet of trump and putin.

well i mean if they use all those zillions of gigawattz for agi there will be no energy left for cancer so they'll starve it

you can't seriously convince anyone over 15 they didn't anticipate this 'a priori', all Amodei, Askell, Cherny etc seem like very smart people...that's just hype for enterprise customers to pay more $$$ or for their upcoming ipo stock.

turn mercury to gold?

hmmm 10x, i added this line on top in the comment on top inside the file:

"PolyForm Noncommercial License 1.0.0"

(free for non-commercial) ^ I'm not sure it's the best license though, actually gpt 5.4 says it's a bad idea, but after a long puzzling wheather to make it GPL free for everything or refrain from posting on github at all i picked the middleground...i may change it later.

either you program AI itself or hardware/robotics: arduino/rasp pi. I have a feeling the next "big thing" will be space but that's a bit premature, make it the mid-30s or 40s where the space-bubble starts.

"free speech" in america...

i've been fan of qwen for quite some time, awsome!

"380K lines, 55+ directories".

" - wait, this is A LOT. For context, gpt-1 was <1000 lines in python (these days on github...), hundreds of times more for more limited domain, it's even suspicious to a point you can argue this is symbolic/statistical engine and not "AI learning". I mean in theory you can add heuristics ad neauseum and generate output, it's not smart per se but likely it would work and it would be O(n) or whatever, it's very surprising especially in high-level language like typescript...if it was assembly at least where most of the job would be working with memory.

personal estimation too, the motivation was to create somethin faster than a transformer since transformers were absurdly slow on my cpu - and it's very obviously faster. I get it llm hype you ad adfinitum regardless...

"Within hours, the ~512,000-line TypeScript codebase was mirrored across GitHub and analyzed by thousands of developers." - kind of weird and cool if claude is in typescript.

"liberate yourself and be satisfied since the one who liberates you is also satisfied"- in the context of Amor Fati and stoicism from "Meditations". Honorable mentions: "Life without meaning is still a life" - the Rebel, Camus and...why not: "Can you hear me John Gordon, can you hear me call?" - from a great work of fiction: Star Kings, E. Hamilton.

it's not beyond the realm of possibility to train it. I emailed 30-40 people already with 37 screenshots including text generation and training. Not a single response. I thought for the time being this path is preferable over huggingspace/github. I'm not ruling out just open source and large .bin pretrained file...i'd need to think about it more.

yeah thanks, for the time being i'm leaning towards 'flash sale' for $1k to a VC. I know it's an absurdly low amount of money for that kind of claim which would lead to suspicions such as: * it's not working at all, * i don't have the code, * the impact is tiny etc. It's like putting a price tag of $1k on a ferrari, you'd think something is definitely wrong. But at the end of the day, it's software: very subjective and fortunately I have couple of other projects that may be received better by the end customer since they're visually more impressive so to speak (better suited for demo videos) so I am willing to sacrifice this one. I've done it in the past frankly: sold for pennies to a Microsoft dev, another one to a DARPA guy and even random redditor once for $2.5k (though this was a "business that makes money" not quite just a working .exe). Like I said 'leaning' towards this, not saying it's the final decision or a good idea at all.

I mean a bit egoistic I guess but can any of these students output such an amazing Chinese song (what im' listening to atm): https://www.youtube.com/watch?v=ugVNDvnDDpA&list=RDugVNDvnDD...

^ Great singer, look up musicians reacting to her (i'm so killing my karma here lol...).

I didn't even click on it and was like 99.999% this will be about "ai version of amazon". Personally: yeah no thanks. I mean I have zero desire to attempt to develop something like this, while I'm sure it will be profitable and whoever makes it will not have to worry about buying a yacht under 50 meters. But when the AGI is knocking on the door, you have to pick your priorities wisely - life's short.

Looks nice, a prima vista. Would love to give it a try later. c#? C# is amazing for GUI apps on windows. A lot of work for single dev still..

No. The exact opposite. It makes me hooked to it even more since now I can finally focus on designing the algorithm instead of thinking about releasing memory, converting datatypes, going to stackoverflow to grab the postgresql connection string or whatever. No more calloc/malloc or int to double or pointers. Plus most of the code generated will compile (ok, "will get interpreted" in the case of Python or whatever you see the point) so if you have designed the right algorithm, it's like 90% job done.

token limitations?? Can OSS 120 billion do it? I noticed even as 100% free version is very generous will give many LOC in codegen.

It's exploding, just look at my post history if you will i just asked about it yesterday that it climbed from 183 to 10th place in the tiobe index in just 5 years! It's absurd, I don't hate it but the 'oject pascal' is just too alien if you come from C, C++ or even C#/java or even PHP background. In fact I don't even know what it resembles? Is it fortran? ok, i googled it: so it's simula, I mean if we pretend object pascal = Delphi and they are very similar or the same thing, right? Is anyone using Simula in 2025? It's not an easy learning path...but I'm not against it, hope it becomes alternative to the "C"-influenced languages.

Great replies yeah. Personally: I'm mid-30s, yet-to-retire addicted to programming, been programming for ages with many highs/lows (aka from developing project with opencv/ai for DARPA to creating the worst flop in history of games which is maybe still tormenting people Steam, i'm not naming it lol it's beyond bad). These days as a hobby: Experimenting with AI, battery designs and propulsion (long story!). I'm based in europe so maybe my English isn't perfect and sounds gpt 3.5 generated, sorry if that's the case.

they're never sending human on Mars.

sizeable minority maybe like 10%-30%, i've seen many posts about Python/ML, i'm very convinced most of the people who focus greatly on Python to do ML are under 30. No source to back it up, just logic - python exploded in the past 5 years and it's hand-in-hand with ml...college kids must be loving it.