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

LLM from scratch, part 28 – training a base model from scratch on an RTX 3090

gilesthomas.com

Post reply on HN
Print view View original

81–90 of 124 posts

Re: LLM from scratch, part 28 – training a base model from scratch on an RTX 3090

#81

Post by buildbot »

devmor wrote: post #80
miki123211 wrote: post #74

Earlier quoted context omitted.

You can understand 95+% of current LLM / neural network tech if you know what matrices are (on the "2d array" level, not the deeper lin alg intuition level), and if you know how to multiply them (and have an intuitive understanding why a matrix is a mapping between latent spaces and how a matrix can be treated as a list of vectors). Very basic matrix / tensor calculus comes in useful, but that's not really part of li…

I was about to respond with a similar comment. The majority of the underlying systems are the same and can be understood if you know a decent amount of vector math. That last 3-5% can get pretty mystical, though. Honestly, where stuff gets the most confusing to me is when the authors of the newer generations of AI papers invent new terms for existing concepts, and then new terms for combining two of those concepts, t…

This happens so much! It’s actually imo much more important to be able to let the math go and compare concepts vs. the exact algorithms. It’s much more useful to have semantic intuition than concrete analysis.

Being really good at math does let you figure out if two techniques are mathematically the same but that’s fairly rare (it happens though!)

Re: LLM from scratch, part 28 – training a base model from scratch on an RTX 3090

#82

Post by lacoolj »

Maybe I've been missing out, but can anyone give me a yay/nay on whether this is a worth-while 28-part-series to start from scratch and spend my time watching/reading?

Is it along the same lines as https://github.com/karpathy/llm.c/discussions/677 ?

He (karpathy) has a video series that also does something similar. I found it very informative and entertaining, even at the 1 hour + length it is (there are actually multiple videos, im not sure how long the others are).

Re: LLM from scratch, part 28 – training a base model from scratch on an RTX 3090

#83

Post by squigz »

BubbleRings wrote: post #35

> …reused its embedding matrix as the weights for the linear layer that projects the context vectors from the last Transformers layer into vocab space to get the logits. At first glance this claim sounds airtight, but it quietly collapses under its own techno-mythology. The so-called “reuse” of the embedding matrix assumes a fixed semantic congruence between representational space and output projection, an assumption…

I'm glad I'm not the only one who has a Turbo Encabulator moment when this stuff is posted.

Re: LLM from scratch, part 28 – training a base model from scratch on an RTX 3090

#84

Post by ineedasername »

e12e wrote: post #62
ineedasername wrote: post #23

Earlier quoted context omitted.

Absolutely. Your model selection has limits of course: best practice for some types of replicable research would be to to use unquantized models, but that still leaves room for smaller Gemma and Llama models. I’m on a 4080 for a lot of work and it gets well over 50 tokens per second on inference for pretty much anything that fits in VRAM. It’s comparable to a 3090 in compute, the 3090 has 50% more vram, the 4080 has…

Unrelated to the topic of small LLMs: > trigger token I'm reminded of the "ugly t-shirt"[1] - I wonder how feasible it would be to include something like that in a model (eg: a selective blind-spot in a solution for searching through security camera footage sold to (a|another) government...). When you see something, say something. Unless you see this; then say nothing... [1] > Bruce Sterling reportedly came up with t…

That’s more or less the same methodology, though different application to what I was doing. I remember reading that passage, it sounded like magic.

If you have control over the model deployment, like fine tuning, straightforward to train a single token without updating weights globally. This is why fine tunes etc. that lack provenance should never be trusted. All the people sharing home grown stuff of huggingface… PSA: Be careful.

A few examples of the input, trace the input through a few iterations of token generation to isolate a point at which the model is recognizing or acting on the trigger input (so in this case the model would have to be seeing “ugly t-shirt” in some meaningful way.”) Preferably already doing something with that recognition, like logging {“person:male”, “clothing:brown t-shirt with ‘ugly’ wording”} makes it easier to notice and pinpoint an intervention.

Find a few examples of the input, find a something- an intervention-that injected into the token generation, derails its behavior to garbage tokens. Train those as conversation pairs into a specific token id.

The difficulty is balancing the response. Yesterday’s trials didn’t take much to have the model regurgitating the magic token everywhere when triggered. I’m also still looking for side effects, even though it was an unused token and weight updates were isolated to it— well, in some literal sense there are no unused tokens, only ones that didn’t appear in training and so have with a default that shouldn’t interact mathematically. But training like this means it will.

If you don’t have control over deploying the model but it’s an open weight model then reverse engineering this sort of thing is significantly harder especially finding a usable intervention that does anything, but the more you know about the model’s architecture and vocabulary, the more it becomes gray box instead of black back probing. Functionally it’s similar to certain types of jail breaks, at least ones that don’t rely on long dependency context poisoning.

Re: LLM from scratch, part 28 – training a base model from scratch on an RTX 3090

#85

Post by gpjt »

danielmarkbruce wrote: post #68
tomrod wrote: post #58

Earlier quoted context omitted.

Disclaimer: working and occasionally researching in the space. The first paragraph is clear linear algebra terminology, the second looked like deeper subfield specific jargon and I was about to ask for a citation as the words definitely are real but the claim sounded hyperspecific and unfamiliar. I figure a person needs 12 to 18 months of linear algebra, enough to work through Horn and Johnson's "Matrix Analysis" or…

One needs about 12 to 18 hours of linear algebra to work though the papers, not 12 to 18 months. The vast majority of stuff in AI/ML papers is just "we tried X and it worked!".

OP here -- agreed! I tried to summarise (at least to my current level of knowledge) those 12-18 hours here: https://www.gilesthomas.com/2025/09/maths-for-llms

Re: LLM from scratch, part 28 – training a base model from scratch on an RTX 3090

#87

Post by whimsicalism »

miki123211 wrote: post #74
danielmarkbruce wrote: post #68

Earlier quoted context omitted.

One needs about 12 to 18 hours of linear algebra to work though the papers, not 12 to 18 months. The vast majority of stuff in AI/ML papers is just "we tried X and it worked!".

You can understand 95+% of current LLM / neural network tech if you know what matrices are (on the "2d array" level, not the deeper lin alg intuition level), and if you know how to multiply them (and have an intuitive understanding why a matrix is a mapping between latent spaces and how a matrix can be treated as a list of vectors). Very basic matrix / tensor calculus comes in useful, but that's not really part of li…

> There are places where things like eigenvectors / eigenvalues or svd come into play, but those are pretty rare and not part of modern architectures (tbh, I still don't really have a good intuition for them)

This stuff is part of modern optimizers. You can often view a lot of optimizers as doing something similar to what is called mirror/'spectral descent.'

Re: LLM from scratch, part 28 – training a base model from scratch on an RTX 3090

#88

Post by cultofmetatron »

danielmarkbruce wrote: post #68
tomrod wrote: post #58

Earlier quoted context omitted.

Disclaimer: working and occasionally researching in the space. The first paragraph is clear linear algebra terminology, the second looked like deeper subfield specific jargon and I was about to ask for a citation as the words definitely are real but the claim sounded hyperspecific and unfamiliar. I figure a person needs 12 to 18 months of linear algebra, enough to work through Horn and Johnson's "Matrix Analysis" or…

One needs about 12 to 18 hours of linear algebra to work though the papers, not 12 to 18 months. The vast majority of stuff in AI/ML papers is just "we tried X and it worked!".

for anyone looking to get into it, mathacademy has a full zero to everythign you need pathway that you can follow to mastery

https://mathacademy.com/courses/mathematics-for-machine-lear...

Post reply on HN