Writing an LLM from scratch, part 6b -- a correction
This is a correction to the sixth in my series of notes on Sebastian Raschka's book "Build a Large Language Model (from Scratch)".
I realised while writing the next part that I'd made a mistake -- while trying to get an intuitive understanding of attention mechanisms, I'd forgotten an important point from the end of my third post. When we convert our tokens into embeddings, we generate two for each one:
- A token embedding that represents the meaning of the token in isolation
- A position embedding that represents where it is in the input sequence.
These two are added element-wise to get an input embedding, which is what is fed into the attention mechanism. However, in my last post I'd forgotten completely about the position embedding and had been talking entirely in terms of token embeddings.
Surprisingly, though, this doesn't actually change very much in that post -- so I've made a few updates there to reflect the change. The most important difference, at least to my mind, is that the fake non-trainable attention mechanism used -- the dot product of the input embeddings -- is, while still excessively basic, not quite as bad as it was. My old example was that in
the fat cat sat on the mat
...the token embeddings for the two "the"s would be the same, so they'd have super-high attention scores for each other. When we consider that it would be the dot product of the input embeddings instead, they'd no longer be identical because they would have different position embeddings. However, the underlying point holds that they would be too closely attending to each other.
Anyway, if you're reading along, I don't think you need to go back and re-read it (unless you particularly want to!). I'm just posting this here for the record :-)
Michael Foord: RIP
Michael Foord, a colleague and friend, passed away this weekend. His passing leaves a huge gap in the Python community.
I first heard from him in early 2006. Some friends and I had just started a new company and there were two of us on the team, both experienced software developers. We'd just hired our third dev, another career coder, but as an XP shop that paired on all production code, we needed a fourth. We posted on the Python.org jobs list to see who we could find, and we got a bunch of applications, among them one from the cryptically-named Fuzzyman, a sales manager at a building supplies merchant who was planning a career change to programming.
He'd been coding as a hobby (I think because a game he enjoyed supported Python scripting), and while he was a bit of an unusual candidate, he wowed us when he came in. But even then, we almost didn't hire him -- there was another person who was also really good, and a bit more conventional, so initially we made an offer to them. To our great fortune, the other person turned the offer down and we asked Michael to join the team. I wrote to my co-founders "it was an extremely close thing and - now that the dust is settling - I think [Michael] may have been the better choice anyway."
That was certainly right! Michael's outgoing and friendly nature changed the company's culture from an inward-facing group of geeks to active members of the UK Python community. He got us sponsoring and attending PyCon UK, and then PyCon US, and (not entirely to our surprise) when we arrived at the conferences, we found that he already appeared to be best friends with everyone. It's entirely possible that he'd never actually met anyone there before -- with Michael, you could never be sure.
Michael's warm-hearted outgoing personality, and his rapidly developing technical skills, made him become an ever-more visible character in the Python community, and he became almost the company's front man. I'm sure a bunch of people only joined our team later because they'd met him first.
I remember him asking one day whether we would consider open-sourcing the rather rudimentary mocking framework we'd built for our internal unit-testing. I was uncertain, and suggested that perhaps he would be better off using it for inspiration while writing his own, better one. He certainly managed to do that.
Sadly things didn't work out with that business, and Michael decided to go his own way in 2009, but we stayed in touch. One of the great things about him was that when you met him after multiple months, or even years, you could pick up again just where you left off. At conferences, if you found yourself without anyone you knew, you could just follow the sound of his booming laugh to know where the fun crowd were hanging out. We kept in touch over Facebook, and I always looked forward to the latest loony posts from Michael Foord, or Michael Fnord as he posted as during his fairly-frequent bans...
This weekend's news came as a terrible shock, and I really feel that we've lost a little bit of the soul of the Python community. Rest in peace, Michael -- the world is a sadder and less wonderfully crazy place without you.
[Update: I was reading through some old emails and spotted that he was telling me I should start blogging in late 2006. So this very blog's existence is probably a direct result of Michael's advice. Please don't hold it against his memory ;-)]
[Update: there's a wonderful thread on discuss.python.org
where people are posting their memories. I highly recommend reading it, and
posting to it if you knew Michael.]
Writing an LLM from scratch, part 6 -- starting to code self-attention
This is the sixth in my series of notes on Sebastian Raschka's book "Build a Large Language Model (from Scratch)". Each time I read part of it, I'm posting about what I found interesting as a way to help get things straight in my own head -- and perhaps to help anyone else that is working through it too. This post covers just one subsection of the trickiest chapter in the book -- subsection 3.3.1, "A simple self-attention mechanism without trainable weights". I feel that there's enough in there to make up a post on its own. For me, it certainly gave me one key intuition that I think is a critical part of how everything fits together.
As always, there may be errors in my understanding below -- I've cross-checked and run the whole post through Claude, ChatGPT o1, and DeepSeek r1, so I'm reasonably confident, but caveat lector :-) With all that said, let's go!
Do reasoning LLMs need their own Philosophical Language?
A few days ago, I saw a cluster of tweets about OpenAI's o1 randomly switching to Chinese while reasoning -- here's a good example. I think I've seen it switch languages a few times as well. Thinking about it, Chinese -- or any other language written in a non-Latin alphabet -- would be particularly noticeable, because those notes describing what it's thinking about flash by pretty quickly, and you're only really likely to notice something weird if it's immediately visibly different to what you expect. So perhaps it's spending a lot of its time switching from language to language depending on what it's thinking about, and then it translates back to the language of the conversation for the final output.
Why would it do that? Presumably certain topics are covered better in its training set in specific languages -- it will have more on Chinese history in Chinese, Russian history in Russian, and so on. But equally possibly, some languages are easier for it to reason about certain topics in. Tiezhen Wang, a bilingual AI developer, tweeted that he preferred doing maths in Chinese "because each digit is just one syllable, which makes calculations crisp and efficient". Perhaps there's something similar there for LLMs.
That got me thinking about the 17th-century idea of a Philosophical Language. If you've read Neal Stephenson's Baroque Cycle books, you'll maybe remember it from there -- that's certainly where I heard about it. The idea was that natural human languages were not very good for reasoning about things, and the solution would be to create an ideal, consciously-designed language that was more rational. Then philosophers (or scientists as we'd say these days) could work in it and get better results.
There are echos of that in E' (E-Prime), another one I picked up on from fiction (this time from The Illuminatus! Trilogy). It's English, without the verb "to be", the idea being that most uses of the word are unnecessarily foggy and would be better replaced. "Mary is a doctor" implies that her job is the important thing about her, whereas "Mary practices medicine" is specific that it's one just aspect of her. What I like about it is that it -- in theory -- gets a more "Philosophical" language with a really small tweak rather than a complete redesign.
What I'm wondering is, are human languages really the right way for LLMs to be reasoning if we want accurate results quickly? We all know how easy it is to be bamboozled by words, either our own or other people's. Is there some way we could construct a language that would be better?
The baroque philosophers ultimately failed, and modern scientists tend to switch to mathematics when they need to be precise ("physics is a system for translating the Universe into maths so that you can reason about it" -- discuss).
But perhaps by watching which languages o1 is choosing for different kinds of reasoning we could identify pre-existing (grammatical/morphological/etc) structures that just seem to work better for different kinds of tasks, and then use that as a framework to build something on top of. That feels like something that could be done much more easily now than it could in the pre-LLM world.
Or maybe a reasoning language is something that could be learned as part of a training process; perhaps each LLM could develop its own, after pre-training with human languages to get it to understand the underlying concept of "language". Then it might better mirror how LLMs work -- its structures might map more directly to the way transformers process information. It might have ways of representing things that you literally could not describe in human languages.
Think of it as a machine code for LLMs, perhaps. Is it a dumb idea? As always, comments are open :-)
Writing an LLM from scratch, part 5 -- more on self-attention
I'm reading Sebastian Raschka's book "Build a Large Language Model (from Scratch)", and posting about what I found interesting every day that I read some of it. In retrospect, it was kind of adorable that I thought I could get it all done over my Christmas break, given that I managed just the first two-and-a-half chapters! However, now that the start-of-year stuff is out of the way at work, hopefully I can continue. And at least the two-week break since my last post in this series has given things some time to stew.
In the last post I was reading about attention mechanisms and how they work, and was a little thrown by the move from attention to self-attention, and in this blog post I hope to get that all fully sorted so that I can move on to the rest of chapter 3, and then the rest of the book. Rashka himself said on X that this chapter "might be the most technical one (like building the engine of a car) but it gets easier from here!" That's reassuring, and hopefully it means that my blog posts will speed up too once I'm done with it.
But first: on to attention and what it means in the LLM sense.
Happy New Year!
A very happy New Year to all for 2025!
Just a quick note to say that I'm not starting one of my periodic blogging holidays. The New Year and then starting work again afterwards has been quite busy, so I've not had much time to continue with my readthrough of Sebastian Raschka's book "Build a Large Language Model (from Scratch)". I was hoping to get some done this weekend, but personal commitments got in the way.
I have had some downtime, but just enough to read some fiction -- I'm rereading Connie Willis's excellent "Oxford Time Travel" series, out of order -- I started with the wonderfully whimsical "To Say Nothing of the Dog", and am now most of the way through the significantly less cheerful but deeply moving "Doomsday Book".
I'm also having a lengthy pair of on-and-off discussions with Claude and ChatGPT o1 about the history of attention mechanisms, so hopefully when I get back to the LLM book I'll be better prepared :-)