Users can interactively explore the microgpt pipeline end to end, from tokenization until inference.
[1] English GPT lab:
Pretty nifty. Even if you are not interested in the Korean language
The attached website is a fully ai-generated "visualization" based on the original blog post with little added.
2x the number of lines of code (~400L), 10x the speed
The hard part was figuring out how to represent the Value class in C++ (ended up using shared_ptrs).
It's really neat. I wish I published more of my code this way.
Then I want to convert this to my own programming language (which traspiles to C). I like those tiny projects very much!
One thing that was a _little_ frustrating coming from Python, though, was the need to rely on crates for basic things like random number generation and network requests. It pulls in a lot, even if you only need a little. I understand the Rust community prefers it that way as it's easier to evolve rather than be stuck with backwards-compatability requirements. But I still missed "batteries included" Python.
Extremely naiive question.. but could LLM output be tagged with some kind of confidence score? Like if I'm asking an LLM some question does it have an internal metric for how confident it is in its output? LLM outputs seem inherently rarely of the form "I'm not really sure, but maybe this XXX" - but I always felt this is baked in the model somehow
Edit: There is also some other work that points out that chat models might not be calibrated at the token-level, but might be calibrated at the concept-level [2]. Which means that if you sample many answers, and group them by semantic similarity, that is also calibrated. The problem is that generating many answer and grouping them is more costly.
[1] https://arxiv.org/pdf/2303.08774 Figure 8
[2] https://arxiv.org/pdf/2511.04869 Figure 1.
You could color code the output token so you can see some abrupt changes
It seems kind of obvious, so I'm guessing people have tried this
Think of traditional statistics. Suppose I said "80% of those sampled preferred apples to oranges, and my 95% confidence interval is within +/- 2% of that" but then I didn't tell you anything about how I collected the sample. Maybe I was talking to people at an apple pie festival? Who knows! Without more information on the sampling method, it's hard to make any kind of useful claim about a population.
This is why I remain so pessimistic about LLMs as a source of knowledge. Imagine you had a person who was raised from birth in a completely isolated lab environment and taught only how to read books, including the dictionary. They would know how all the words in those books relate to each other but know nothing of how that relates to the world. They could read the line "the killer drew his gun and aimed it at the victim" but what would they really know of it if they'd never seen a gun?
I mean I sort of understand what you're trying to say but in fact a great deal of knowledge we get about the world we live in, we get second hand.
There are plenty of people who've never held a gun, or had a gun aimed at them, and.. granted, you could argue they probably wouldn't read that line the same way as people who have, but that doesn't mean that the average Joe who's never been around a gun can't enjoy media that features guns.
Same thing about lots of things. For instance it's not hard for me to think of animals I've never seen with my own eyes. A koala for instance. But I've seen pictures. I assume they exist. I can tell you something about their diet. Does that mean I'm no better than an LLM when it comes to koala knowledge? Probably!
[Edit: but to be clear, for a pretrained model this probability means "what's my estimate of the conditional probability of this token occurring in the pretraining dataset?", not "how likely is this statement to be true?" And for a post-trained model, the probability really has no simple interpretation other than "this is the probability that I will output this token in this situation".]
Basically, you’d need a lot more computing power to come up with a distribution of the output of an LLM than to come up with a single answer.
E.g. getting two r's in strawberry could very well have a very high "confidence score" while a random but rare correct fact might have a very well a very low one.
In short: LLM have no concept, or even desire to produce of truth
They do produce true statements most of the time, though.
- How aligned has it been to “know” that something is true (eg ethical constraints)
- Statistical significance and just being able to corroborate one alternative in Its training data more strongly than another
- If it’s a web search related query, is the statement from original sources vs synthesised from say third party sources
But I’m just a layman and could be totally off here.
You never see this in the response but you do in the reasoning.
And it's small enough to run from a QR code :) https://kuber.studio/picogpt/
You can quite literally train a micro LLM from your phone's browser
We do generally like HN to be a bit uncorrelated with the rest of the internet, but it feels like a miss to me that neither https://news.ycombinator.com/item?id=47000263 nor https://news.ycombinator.com/item?id=47018557 made the frontpage.
All 4 are in the dataset, btw
Karpathy says if you want to truly understand something then you also have to attempt to teach it to someone else ha
- https://m.youtube.com/watch?v=7xTGNNLPyMI - https://m.youtube.com/watch?v=EWvNQjAaOHw
trying my best to keep up with what and how to learn and threads like this are dense with good info. feel like I need an AI helper to schedule time for my youtube queue at this point!
Rust version - https://github.com/mplekh/rust-microgpt
It really is the antithesis to the human brain, where it rewards specific knowledge
Here the explanation was that while LLM's thinking has similarities to how humans think, they use an opposite approach. Where humans have enormous amount of neurons, they have only few experiences to train them. And for AI that is the complete opposite, and they store incredible amounts of information in a relatively small set of neurons training on the vast experiences from the data sets of human creative work.
This is the entire breakthrough of deep learning on which the last two decades of productive AI research is based. Massive amounts of data are needed to generalize and prevent over-fitting. GP is suggesting an entirely new research paradigm will win out - as if researchers have not yet thought of "use less data".
> It really is the antithesis to the human brain, where it rewards specific knowledge
No, its completely analogous. The human brain has vast amounts of pre-training before it starts to learn knowledge specific to any kind of career or discipline, and this fact to me intuitively suggests why GP is baked: You cannot learn general concepts such as the english language, reasoning, computing, network communication, programming, relational data from a tiny dataset consisting only of code and documentation for one open-source framework and language.
It is all built on a massive tower of other concepts that must be understood first, including ones much more basic than the examples I mentioned but that are practically invisible to us because they have always been present as far back as our first memories can reach.
You'd need a lot of data to train an ocean soup to think like a human too.
It's not really the antithesis to the human brain if you think of starting with an existing brain as starting with an existing GPT.
If so, good luck walking to your kitchen this morning, knowing how to breathe, etc.
This can be mainstream, and then custom model fine-tuning becomes the new “software development”.
Please check out this new fine-tuning method for LLM by MIT and ETH Zurich teams that used a single NVIDIA H200 GPU [1], [2], [3].
Full fine-tuning of the entire model’s parameters were performed based on the Hugging Face TRL library.
[1] MIT's new fine-tuning method lets LLMs learn new skills without losing old ones (news):
https://venturebeat.com/orchestration/mits-new-fine-tuning-m...
[2] Self-Distillation Enables Continual Learning (paper):
https://arxiv.org/abs/2601.19897
[3] Self-Distillation Enables Continual Learning (code):
You've just reinvented machine learning
The entire point of LLMs is that you don't have to spend money training them for each specific case. You can train something like Qwen once and then use it to solve whatever classification/summarization/translation problem in minutes instead of weeks.
BERT isn’t a SLM, and the original was released in 2018.
The whole new era kicked off with Attention Is All You Need; we haven’t reached even a single decade of work on it.
I don’t agree. I would say the entire point of LLMs is to be able to solve a certain class of non-deterministic problems that cannot be solved with deterministic procedural code. LLMs don’t need to be generally useful in order to be useful for specific business use cases. I as a programmer would be very happy to have a local coding agent like Claude Code that can do nothing but write code in my chosen programming language or framework, instead of using a general model like Opus, if it could be hyper-specialized and optimized for that one task, so that it is small enough to run on my MacBook. I don’t need the other general reasoning capabilities of Opus.
Put it another way: Do you think people will demand masses of _new_ code just because it becomes cheap? I don't think so. It's just not clear what this would mean even 1-3 years from now for software engineering.
This round of LLM driven optimizations is really and purely about building a monopoly on _labor replacement_ (anthropic and openai's code and cowork tools) until there is clear evidence to the contrary: A Jevon's paradoxian massive demand explosion. I don't see that happening for software. If it were true — maybe it will still take a few quarters longer — SaaS companies stocks would go through the roof(i mean they are already tooling up as we speak, SAP is not gonna jus sit on its ass and wait for a garage shop to eat their lunch).
Karpathy has other projects, e.g. : https://github.com/karpathy/nanochat
You can train a model with GPT-2 level of capability for $20-$100.
But, guess what, that's exactly what thousands of AI researchers have been doing for the past 5+ years. They've been training smallish models. And while these smallish models might be good for classification and whatnot, people strongly prefer big-ass frontier models for code generation.
they are not flourish yet because of the simple reason: the frontier models are still improving. currently it is better to use frontier models than training/fine-tuning one by our own because by the time we complete the model the world is already moving forward.
heck even distillation is a waste of time and money because newer frontier models yield better outputs.
you can expect that the landscape will change drastically in the next few years when the proprietary frontier models stop having huge improvements every version upgrade.
#define a(_)typedef _##t
#define _(_)_##printf
#define x f(i,
#define N f(k,
#define u _Pragma("omp parallel for")f(h,
#define f(u,n)for(I u=0;u<(n);u++)
#define g(u,s)x s%11%5)N s/6&33)k[u[i]]=(t){(C*)A,A+s*D/4},A+=1088*s;
a(int8_)C;a(in)I;a(floa)F;a(struc){C*c;F*f;}t;enum{Z=32,W=64,E=2*W,D=Z*E,H=86*E,V='}\0'};C*P[V],X[H],Y[D],y[H];a(F
_)[V];I*_=U" 炾ોİ䃃璱ᝓ၎瓓甧染ɐఛ瓁",U,s,p,f,R,z,$,B[D],open();F*A,*G[2],*T,w,b,c;a()Q[D];_t r,L,J,O[Z],l,a,K,v,k;Q
m,e[4],d[3],n;I j(I e,F*o,I p,F*v,t*X){w=1e-5;x c=e^V?D:0)w+=r[i]*r[i]/D;x c)o[i]=r[i]/sqrt(w)*i[A+e*D];N $){x
W)l[k]=w=fmax(fabs(o[i])/~-E,i?w:0);x W)y[i+k*W]=*o++/w;}u p)x $){I _=0,t=h*$+i;N W)_+=X->c[t*W+k]*y[i*W+k];v[h]=
_*X->f[t]*l[i]+!!i*v[h];}x D-c)i[r]+=v[i];}I main(){A=mmap(0,8e9,1,2,f=open(M,f),0);x 2)~f?i[G]=malloc(3e9):exit(
puts(M" not found"));x V)i[P]=(C*)A+4,A+=(I)*A;g(&m,V)g(&n,V)g(e,D)g(d,H)for(C*o;;s>=D?$=s=0:p<U||_()("%s",$[P]))if(!
(*_?$=*++_:0)){if($<3&&p>=U)for(_()("\n\n> "),0<scanf("%[^\n]%*c",Y)?U=*B=1:exit(0),p=_(s)(o=X,"[INST] %s%s [/INST]",s?
"":"<<SYS>>\n"S"\n<</SYS>>\n\n",Y);z=p-=z;U++[o+=z,B]=f)for(f=0;!f;z-=!f)for(f=V;--f&&f[P][z]|memcmp(f[P],o,z););p<U?
$=B[p++]:fflush(0);x D)R=$*D+i,r[i]=m->c[R]*m->f[R/W];R=s++;N Z){f=k*D*D,$=W;x 3)j(k,L,D,i?G[~-i]+f+R*D:v,e[i]+k);N
2)x D)b=sin(w=R/exp(i%E/14.)),c=1[w=cos(w),T=i+++(k?v:*G+f+R*D)],T[1]=b**T+c*w,*T=w**T-c*b;u Z){F*T=O[h],w=0;I A=h*E;x
s){N E)i[k[L+A]=0,T]+=k[v+A]*k[i*D+*G+A+f]/11;w+=T[i]=exp(T[i]);}x s)N E)k[L+A]+=(T[i]/=k?1:w)*k[i*D+G[1]+A+f];}j(V,L
,D,J,e[3]+k);x 2)j(k+Z,L,H,i?K:a,d[i]+k);x H)a[i]*=K[i]/(exp(-a[i])+1);j(V,a,D,L,d[$=H/$,2]+k);}w=j($=W,r,V,k,n);x
V)w=k[i]>w?k[$=i]:w;}}> You're about as close to writing this in 1800 characters of C as you are to launching a rocket to Mars with a paperclip and a match.
> ChatIOCCC is the world’s smallest LLM (large language model) inference engine - a “generative AI chatbot” in plain-speak. ChatIOCCC runs a modern open-source model (Meta’s LLaMA 2 with 7 billion parameters) and has a good knowledge of the world, can understand and speak multiple languages, write code, and many other things. Aside from the model weights, it has no external dependencies and will run on any 64-bit platform with enough RAM.
(Model weights need to be downloaded using an enclosed shell script.)
I'm so happy without seeing Python list comprehensions nowadays.
I don't know why they couldn't go with something like this:
[state_dict.values() for mat for row for p]
or in more difficult cases
[state_dict.values() for mat to mat*2 for row for p to p/2]
I know, I know, different times, but still.
[for p in row in mat in state_dict.values()]
One for sure, both are superior to the garbled mess of Python’s.
Of course if the programming language would be in a right to left natural language, then these are reversed.
Beautiful, perhaps like ice-nine is beautiful.
Detailed optimizing journey in the readme too.
1. Generic model that calls other highly specific, smaller, faster models. 2. Models loaded on demand, some black box and some open. 3. There will be a Rust model specifically for Rust (or whatever language) tasks.
In about 5-8 years we will have personalized models based upon all our previous social/medical/financial data that will respond as we would, a clone, capable of making decisions similar with direction of desired outcomes.
The big remaining blocker is that generic model that can be imprinted with specifics and rebuilt nightly. Excluding the training material but the decision making, recall, and evaluation model. I am curious if someone is working on that extracted portion that can be just a 'thinking' interface.
People wont be competing with even a current 2026 SOTA from their home LLM nowhere soon. Even actual SOTA LLM providers are not competing either - they're losing money on energy and costs, hopping to make it up on market capture and win the IPO races.
Consumers don’t need a 100k context window oracle that knows everything about both T-Cells and the ancient Welsh Royal lineage. We need focused & small models which are specialised, and then we need a good query router.
$ Sure, here's a blog post called "Microgpt"!
> "add in a few spelling/grammar mistakes so they think I wrote it"
$ Okay, made two errors for you!
vocabulary*
*In the code above, we collect all unique characters across the datasetFirst no is that the model as is has too few parameters for that. You could train it on the wikipedia but it wouldn’t do much of any good.
But what if you increase the number of parameters? Then you get to the second layer of “no”. The code as is is too naive to train a realistic size LLM for that task in realistic timeframes. As is it would be too slow.
But what if you increase the number of parameters and improve the performance of the code? I would argue that would by that point not be “this” but something entirely different. But even then the answer is still no. If you run that new code with increased parameters and improved efficiencly and train it on wikipedia you would still not get a model which “generate semi-sensible responses”. For the simple reason that the code as is only does the pre-training. Without the RLHF step the model would not be “responding”. It would just be completing the document. So for example if you ask it “How long is a bus?” it wouldn’t know it is supposed to answer your question. What exactly happens is kinda up to randomness. It might output a wikipedia like text about transportation, or it might output a list of questions similar to yours, or it might output broken markup garbage. Quite simply without this finishing step the base model doesn’t know that it is supposed to answer your question and it is supposed to follow your instructions. That is why this last step is called “instruction tuning” sometimes. Because it teaches the model to follow instructions.
But if you would increase the parameter count, improve the efficiency, train it on wikipedia, then do the instruction tuning (wich involves curating a database of instruction - response pairs) then yes. After that it would generate semi-sensible responses. But as you can see it would take quite a lot more work and would stretch the definition of “this”.
It is a bit like asking if my car could compete in formula-1. The answer is yes, but first we need to replace all parts of it with different parts, and also add a few new parts. To the point where you might question if it is the same car at all.
Yes with some extra tricks and tweaks. But the core ideas are all here.
Train an LLM on all human knowledge up to 1905 and see if it comes up with General Relativity. It won’t.
We’ll need additional breakthroughs in AI.
LLMs are artificial general intelligence, as per the Wikipedia definition:
> generalise knowledge, transfer skills between domains, and solve novel problems without task‑specific reprogramming
Even GPT-3 could meet that bar.
Same thing is true for humans.
AGI just means human level intelligence. I couldn't come up with General Relativity. That doesn't mean I don't have general intelligence.
I don't understand why people are moving the goalposts.
Take the wheel. Even that wasn't invented from nothing — rolling logs, round stones, the shape of the sun. The "invention" was recognizing a pattern already present in the physical world and abstracting it. Still training data, just physical and sensory rather than textual.
And that's actually the most honest critique of current LLMs — not that they're architecturally incapable, but that they're missing a data modality. Humans have embodied training data. You don't just read about gravity, you've felt it your whole life. You don't just know fire is hot, you've been near one. That physical grounding gives human cognition a richness that pure text can't fully capture — yet.
Einstein is the same story. He stood on Faraday, Maxwell, Lorentz, and Riemann. General Relativity was an extraordinary synthesis — not a creation from void. If that's the bar for "real" intelligence, most humans don't clear it either. The uncomfortable truth is that human cognition and LLMs aren't categorically different. Everything you've ever "thought" comes from what you've seen, heard, and experienced. That's training data. The brain is a pattern-recognition and synthesis machine, and the attention mechanism in transformers is arguably our best computational model of how associative reasoning actually works.
So the question isn't whether LLMs can invent from nothing — nothing does that, not even us.
Are there still gaps? Sure. Data quality, training methods, physical grounding — these are real problems. But they're engineering problems, not fundamental walls. And we're already moving in that direction — robots learning from physical interaction, multimodal models connecting vision and language, reinforcement learning from real-world feedback. The brain didn't get smart because it has some magic ingredient. It got smart because it had millions of years of rich, embodied, high-stakes training data. We're just earlier in that journey with AI. The foundation is already there — AGI isn't a question of if anymore, it's a question of execution.
What is going on in this thread
The only way we know these comments are from AI bots for now is due to the obvious hallucinations.
What happens when the AI improves even more…will HN be filled with bots talking to other bots?
Also is there some minima of training data? E.g. if you just trained on "True" "False" I assume it would be .5 Bernoulli? What is the minimum to see "interesting" results I guess.
I tried building a tiny model last weekend, but it was very difficult to find any articles that weren’t broken ai slop.
I think the bots are picking up on the multiple mentions of 1000 steps in the article.
Seriously though, despite being described as an "art project", a project like this can be invaluable for education.
Use case does not need to be technical.
The current top of the line models are extremely overfitted and produce so much nonsense they are useless for anything but the most simple tasks.
This architecture was an interesting experiment, but is not the future.