Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Microgpt

karpathy.github.io

Post reply on HN
Print view View original

161–170 of 354 posts

Re: Microgpt

#161

Post by npn »

znnajdla wrote: post #105
npn wrote: post #101

Earlier quoted context omitted.

what gut? we are already doing that. there are a lot of "tiny" LLMs that are useful: M$ Phi-4, Gemma 3/3n, Qwen 7B... There are even smaller models like Gemma 270M that is fine tuned for function calls. 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 t…

I’ve tried those tiny LLMs and they don’t seem useful to me for real world tasks. They are toys for super simple autocomplete.

Did you use them as-is, or had you fine tuned them for your particular use cases?

Re: Microgpt

#162

Post by mentos »

la_fayette wrote: post #157

This guy is so amazing! With his video and the code base I really have the feeling I understand gradient descent, back propagation, chain rule etc. Reading math only just confuses me, together with the code it makes it so clear! It feels like a lifetime achievement for me :-)

Curious if you could try to explain it. It’s my goal to sit down with it and attempt to understand it intuitively.

Karpathy says if you want to truly understand something then you also have to attempt to teach it to someone else ha

Re: Microgpt

#163

Post by sharperguy »

Lionga wrote: post #146
geokon wrote: post #141

> What’s the deal with “hallucinations”? The model generates tokens by sampling from a probability distribution. It has no concept of truth, it only knows what sequences are statistically plausible given the training data. 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…

The LLM has an internal "confidence score" but that has NOTHING to do with how correct the answer is, only with how often the same words came together in training data. 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

Still, it might be interesting information to have access to, as someone running the model? Normally we are reading the output trying to build an intuition for the kinds of patterns it outputs when it's hallucinating vs creating something that happens to align with reality. Adding in this could just help with that even when it isn't always correlated to reality itself.

Re: Microgpt

#165

Post by spinningslate »

growingswe wrote: post #98

Great stuff! I wrote an interactive blogpost that walks through the code and visualizes it: https://growingswe.com/blog/microgpt

That’s beautifully done, thanks for posting. As helpful again to an ML novice like me as Karpathy’s original.

Re: Microgpt

#166

Post by throw310822 »

aix1 wrote: post #82
thatxliner wrote: post #78

Earlier quoted context omitted.

wiat what does this do?

As the contest entry page explains: > 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 depende…

Good reminder of the fact that an LLM is not a program.

Re: Microgpt

#167

Post by systima »

znnajdla wrote: post #95

Super useful exercise. My gut tells me that someone will soon figure out how to build micro-LLMs for specialized tasks that have real-world value, and then training LLMs won’t just be for billion dollar companies. Imagine, for example, a hyper-focused model for a specific programming framework (e.g. Laravel, Django, NextJS) trained only on open-source repositories and documentation and carefully optimized with a spec…

[dead]

Re: Microgpt

#168

Post by bitwize »

the_af wrote: post #87
thatxliner wrote: post #77

Earlier quoted context omitted.

btw my friend is asking if your username is a "Klara and the Sun" reference

I've read the book and I'm a fan of Ishiguro in general, but I'm failing to make the reference, so I'm going to go with "no" :)

The robots in the book were called Artificial Friends, or AFs.

Re: Microgpt

#169

Post by la_fayette »

mentos wrote: post #162
la_fayette wrote: post #157

This guy is so amazing! With his video and the code base I really have the feeling I understand gradient descent, back propagation, chain rule etc. Reading math only just confuses me, together with the code it makes it so clear! It feels like a lifetime achievement for me :-)

Curious if you could try to explain it. It’s my goal to sit down with it and attempt to understand it intuitively. Karpathy says if you want to truly understand something then you also have to attempt to teach it to someone else ha

Yes, that’s true! That could be my next step… though I have to admit, writing this in a HN comment feels like a bit of a challenge.

Re: Microgpt

#170

Post by krige »

Dansvidania wrote: post #110
jimbokun wrote: post #85

Earlier quoted context omitted.

No 16 year old has practiced driving a car for 16 years.

If you see gaining fine motor control, understanding pictographic language […] as a prerequisite to driving a car, then yes, all of them are

That's an exaggeration. Nobody is trained to read STOP signs for 16 years, a few months top. And Waymo doesn't need to coordinate a four-limbed, 20-digited, one-headed body to operate a car.
Post reply on HN