Doesn't this indicate that there's not a lot of transfer learning between languages? If the networks read a prompt, reason about it and then present the result, only the reading and writing part would differ between languages, and that should be a small part of the full energy use.

It would be interesting to see if it would make any difference if you posttrained with translations of the same texts, i.e. focusing on teaching the model to specifically translate concepts between languages.

▲

ozgung 5 hours ago | parent | next [-]

My understanding is that it’s about tokenization. English input is converted into less tokens because tokenizer is optimized for English. This makes huge difference because all the output tokens are fed back into the input at each token generation. And attention blocks are generally quadratic with the input tokens. So foreign language means more tokens everywhere and more tokens mean more passes and much more compute and memory.

▲

not_a_bot_4sho 4 hours ago | parent | next [-]

This is absolutely correct.

But I would subtly add it's also about the training data, which is optimized towards English. Not intentionally, just incidentally. (And this is a really hot area of improvement with a long way to go.)

Which is largely why the tokenization part that you mentioned is what it is.

▲

everforward 4 hours ago | parent | prev | next [-]

My understanding (which may be wrong) is that English is best but most of the Romance languages do okay because English has borrowed from them so much.

I think you only start getting seriously bad impacts in non-Romance languages where they have their own alphabet, or use the English alphabet phonetically causing letters to appear next to each other that never would in English. E.g. in English "qu" is a pretty obvious token, you rarely see a Q without a U after it. Happens all the time in anglicized Middle Eastern words, though. Quran is like a whole other thing, you'd almost always expect another vowel after qu, not a consonant.

	▲	poopingatwork 3 hours ago \| parent [-]
		I've heard English is one of the most expressive languages, but is complex to learn due to absorbing words and ideas from so many other languages. Lots of contradictions, situational rules, and unwritten rules (like the order of adjectives: the big, brown chair instead of the brown, big chair). I don't know how true that is, and I can't imagine knowing more than one language. I'm impressed by anyone who knows more than one....

▲

sorenjan 4 hours ago | parent | prev [-]

Can't you train a model to translate tokens from human languages to a separate NN token language with a compressed representation like a VAE? It seems like a leaky abstraction to have these trillion parameter models be dependant on the fact that English gets tokenized in a certain way, and I assume this has been researched a lot already.

▲

breakwaterlabs 5 hours ago | parent | prev [-]

"Reason about it" is where your abstraction has sprung a leak.

LLMs do what they say on the tin: they model language. If you ask an LLM "find security vulnerabilities in my code" for hello_world.rs, the next likely tokens are "I found the following vulnerabilities", regardless of what logic or reason say.

It happens to be true that predicting next tokens on your own output mocks actual reasoning fairly well. But it still just a simulacrum rather than the real thing, and when held as an abstraction it leaks everywhere.

Understanding this is critical to avoiding the common failure case of scope creep and code slop.

	▲	sorenjan 4 hours ago \| parent \| next [-]
		I only used "reason about" as shorthand for the computation that takes place between input text gets tokenized and output tokens gets translated back to text. There are a lot of layers in the middle where the tokens no longer represent text directly but rather a combination of the different tokens through the attention and MLP projections.
	▲	elcritch 4 hours ago \| parent \| prev \| next [-]
		There’s enough experience with LLMs now and comparisons to our own thinking that we can say LLMs are thinking and doing so not so differently from us - at least for some aspects of reasoning. Trying to pretend otherwise at this point doesn’t help us to utilize them or understand their limits. We’re better off accepting that intelligence isn’t unique to us as a species. Then we can see that modeling language as a tool for us as a vehicle for thinking and reasoning, which philosophers like Wittgenstein already landed upon before LLMs.
	▲	wat10000 2 hours ago \| parent \| prev [-]
		No decent modern LLM is going to start its reply with “I found the following vulnerabilities” when you ask it to find security vulnerabilities in a straightforward hello world.