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

We Found an Neuron in GPT-2

clementneo.com

Post reply on HN
Print view View original

21–30 of 177 posts

Re: We Found an Neuron in GPT-2

#21

Post by kelnos »

tqi wrote: post #10

> We started out with the question: How does GPT-2 know when to use the word an over a? The choice depends on whether the word that comes after starts with a vowel or not, but GPT-2 is only capable of predicting one word at a time. We still don’t have a full answer... I'm not sure I understand why this is an open question. While I get that GPT-2 is predicting only one word at a time, it doesn't seem that surprising t…

Yeah, my feeling here was that it's sort of tautological: if GPT predicts "a", then it must then predict a word that would follow "a" and not require "an" (and vice versa). And if you think about it from the opposite direction: if it's working out a response that is eventually going to have "apple" in it, then all the data it's trained on is going to cause it to predict "an" even before it needs to predict "apple".

(Admittedly, all this ML/AI stuff is still beyond my current level of understanding, so I'm sure my thinking here is off.)

Re: We Found an Neuron in GPT-2

#22

Post by RobotToaster »

fragmede wrote: post #14
unshavedyak wrote: post #6

I wonder if this stuff will ever be applicable to a person and a laptop (or if it is now?). Ie this seems like such a cool area to be in but the data volumes required are huge, complex, etc. Code is simple, cheap, lean, etc by comparison. Do we have any insight on how this area of research could be usable with less hardware and data? Is there a visible future where a guy and a laptop can make a big program? (without…

Does this hypothetical laptop have a GPU? StableDiffusion is in this realm of "stuff" and is runnable on consumer GPU systems. It's a bit of trouble to get setup if you're not a python dev (and kinda still is if you are) but it's a pretty neat ML model to play around with.

My 2 year old gaming laptop runs stablediffusion just fine. Only having a gtx 1660 ti 6gb it's a little at the limit of what will run it though.

Re: We Found an Neuron in GPT-2

#23

Post by m3kw9 »

Grammatically all words depend on each other in close sequence. “The sun was shining” vs “it’s an apple”. It doesn’t know each words probability independently, it already knows the entire sequence’s probability before hand from the model

Re: We Found an Neuron in GPT-2

#24

Post by frogulis »

edgyquant wrote: post #19
jakear wrote: post #4

Earlier quoted context omitted.

More accurately, it's whether the word that comes after starts with a vowel sound . This is why `an 'istoric` is correct, and `a historic` is correct, but `an historic` is incorrect (as famously used by Steven Colbert).

I is a vowel so your rule about sound doesn’t apply here. It’s also not a rule I’ve ever heard, in school we’re taught that only vowels get a before them

> in school we’re taught that only vowels get a before them

That's the opposite of how it works (as you demonstrated in your own comment, "not a rule...")

The comment you responded to was correct, or at least very reasonable, and I'm really not sure which part you disagree with.

Re: We Found an Neuron in GPT-2

#25

Post by rcoveson »

jakear wrote: post #4
nailer wrote: post #3

> The choice depends on whether the word that comes after starts with a vowel or not Or an aitch / h, but only sometimes.

More accurately, it's whether the word that comes after starts with a vowel sound . This is why `an 'istoric` is correct, and `a historic` is correct, but `an historic` is incorrect (as famously used by Steven Colbert).

I'm always thrown when people write "an SQL parser". I'll bet roughly half this crowd is the opposite and cringe when they read "a SQL parser".

Re: We Found an Neuron in GPT-2

#26

Post by Tao3300 »

jakear wrote: post #4
nailer wrote: post #3

> The choice depends on whether the word that comes after starts with a vowel or not Or an aitch / h, but only sometimes.

More accurately, it's whether the word that comes after starts with a vowel sound . This is why `an 'istoric` is correct, and `a historic` is correct, but `an historic` is incorrect (as famously used by Steven Colbert).

Canst thou draw out leviathan with an hook?

Re: We Found an Neuron in GPT-2

#27

Post by Sharlin »

laszlokorte wrote: post #18
tqi wrote: post #10

> We started out with the question: How does GPT-2 know when to use the word an over a? The choice depends on whether the word that comes after starts with a vowel or not, but GPT-2 is only capable of predicting one word at a time. We still don’t have a full answer... I'm not sure I understand why this is an open question. While I get that GPT-2 is predicting only one word at a time, it doesn't seem that surprising t…

From my naïve point of view it seems obvious that at any point where both an „a“ or a „an“ would fit the model randomly selects one of them and by doing so reduces the set of possible nouns to follow.

No, of course not randomly. If you RTA, it should be clear that some prompts are more likely to be followed by an “an” than others.

Re: We Found an Neuron in GPT-2

#28

Post by fragmede »

edgyquant wrote: post #19
jakear wrote: post #4

Earlier quoted context omitted.

More accurately, it's whether the word that comes after starts with a vowel sound . This is why `an 'istoric` is correct, and `a historic` is correct, but `an historic` is incorrect (as famously used by Steven Colbert).

I is a vowel so your rule about sound doesn’t apply here. It’s also not a rule I’ve ever heard, in school we’re taught that only vowels get a before them

https://www.ecenglish.com/learnenglish/lessons/vowel-sounds-...

"Vowel Sounds"

Re: We Found an Neuron in GPT-2

#29

Post by Tao3300 »

rcoveson wrote: post #25
jakear wrote: post #4

Earlier quoted context omitted.

More accurately, it's whether the word that comes after starts with a vowel sound . This is why `an 'istoric` is correct, and `a historic` is correct, but `an historic` is incorrect (as famously used by Steven Colbert).

I'm always thrown when people write "an SQL parser". I'll bet roughly half this crowd is the opposite and cringe when they read "a SQL parser".

I'm in the special superstate where both make me cringe and wish there was a better way!

Solution get rid of article. Someone want to know if thing is definitive thing, can figure it out from context!

Re: We Found an Neuron in GPT-2

#30

Post by PeterisP »

LudwigNagasena wrote: post #13

It’s notable how successful LLMs despite the lack of any linguistic tools in their architectures. It would be interesting to know how different a model would be if it operated on eg dependency trees instead of the linear list of tokens. Surely, the question of “a/an” would be solved with ease as the model would be required to come up with a noun token before choosing its determiner. I wonder if the developers of LLMs…

I think the lack of explicit linguistic tools is the key to success, forcing/enabling the generic model to learn implicit linguistic tools (there's some research identifying that analysis of specific linguistic phenomena happens at specific places in the NN layers) that work better than what we could implement.

"It would be interesting to know how different a model would be if it operated on eg dependency trees instead of the linear list of tokens." - indeed, this is obviously interesting, so people have tried that a lot for many models, but IMHO it's probably now almost decade since the consensus is that in general end-to-end training (once we became able to do it) work better than adding explicit stages in between, e.g. for any random task I would expect that doing text->syntax tree->outcome is going to get worse results than text->outcome, because even if the task really needs syntax, the syntax representation that a stack of large transformer layers learns implicitly tends to be a better representation of the natural language than any human linguist devised formal grammar, which inevitably has to mangle the actual languge to fit into neat human-analyzable 'boxes'/classes/types in which it doesn't really fit and all the fuzzy edge cases stick out. Once you remove the constraint that the grammar must be simplified enough for a human to be able to rationally analyze and understand, and (perhaps even more importantly?) abandoning the need to prematurely disambiguate the utterance to a single syntax tree instead of accepting that parts of it are ambiguous (but not equally likely), processing works better.

It's just one more reminder of the bitter lesson (http://incompleteideas.net/IncIdeas/BitterLesson.html) which we don't want to accept.

Post reply on HN