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

People are just as bad as my LLMs

wilsoniumite.com

Post reply on HN
Print view View original

31–40 of 173 posts

Re: People are just as bad as my LLMs

#31

Post by thfuran »

da_chicken wrote: post #13
aidos wrote: post #8

Earlier quoted context omitted.

Why is that ? Whenever I’m giving examples I almost always use 7, something ending in a 7 or something in the 70s

The theory I've heard is that the more prime a number is, the more random it feels. 13 feels more awkward and weird, and it doesn't come up naturally as often as 2 or 3 do in everyday life. It's rare, so it must be more random! I'll give you the most random number I can think of! People tend to avoid extremes, too. If you ask for a number between 1 and 10, people tend to pick something in the middle. Somehow, the ord…

People also tend to botch random sequences by trying to avoid repetition or patterns.

Re: People are just as bad as my LLMs

#32

Post by mynameismon »

tehsauce wrote: post #3

There has been some good research published on this topic of how RLHF, ie aligning to human preferences easily introduces mode collapse and bias into models. For example, with a prompt like: "Choose a random number", the base pretrained model can give relatively random answers, but after fine tuning to produce responses humans like, they become very biased towards responding with numbers like "7" or "42".

Can you share any links about this?

Re: People are just as bad as my LLMs

#33

Post by taormina »

Jensson wrote: post #29
d4mi3n wrote: post #11

Earlier quoted context omitted.

My guess is that we bias towards numbers with cultural or personal significance. 7 is lucky in western cultures and is religiously significant (see https://en.wikipedia.org/wiki/7#Culture ). 42 is culturally significant in science fiction, though that's a lot more recent. There are probably other examples, but I imagine the mean converges on numbers with multiple cultural touchpoints.

I have never heard of 7 being a lucky number in western culture and your link doesn't support that. 3 is a lucky number, 13 is an unlucky number, 7 is nothing to me. So I don't think its that, 7 is still a very common "random number" here even though there is no special cultural significance to it.

Have you heard of Las Vegas? The 777 being the grand prize? Maybe it is not universal to all of western society but I have never before today heard of a culture where 3 was the lucky number. The USA’s culturally lucky number is absolutely 7.

Re: People are just as bad as my LLMs

#35

Post by K0balt »

I know this is only adjacent to OP’s point, but I do find it somewhat ironic that it is easy to find people who are just as unreliable and incompetent at answering questions correctly as a 7b model, but also a lot less knowledgeable.

Also, often less capable of carrying on a decent conversation.

I’ve noticed an periconcious urge when talking to people to judge them against various models and quants, or to decide they are truly SOTA.

I need to touch grass a bit more, I think.

Re: People are just as bad as my LLMs

#36

Post by rahimnathwani »

harrisonjackson wrote: post #18
devit wrote: post #4

The "person one" vs "person two" bias seems trivially solvable by running each pair evaluation twice with each possible labelling and the averaging the scores. Although of course that behavior may be a signal that the model is sort of guessing randomly rather than actually producing a signal.

Agreed on the second part. Correcting for bias this way might average out the scores but not in a way that correctly evaluates the HN comments. The LLM isn't performing the desired task. It sounds possible to cancel out the comments where reversing the labels swaps the outcome because of bias. That will leave the more "extreme" HN comments that it consistently scored regardless of the label. But that may not solve fo…

  The LLM isn't performing the desired task.
It's 'not performing the task', in the same way that the humans ranking voice attractiveness are 'not performing the task'.

I wouldn't treat the output as complete garbage, just because it's somewhat biased by an irrelevant signal.

Re: People are just as bad as my LLMs

#37

Post by consumer451 »

th0ma5 wrote: post #14
mdp2021 wrote: post #9

Very nice article. But the title, and the idea, is the very frequent "racist" form of the proper "People [can be] just as bad as my LLMs". Now: some people can't count. Some people hum between words. Some people set fire to national monuments. Reply: "Yes we knew", and "No, it's not necessary". And: if people could lift the tons, we would not have invented cranes. Very, very often in these pages I meet people repeati…

Racist is the wrong word probably maybe ... antisocial in that it is against society.

Maybe misanthropic?

Re: People are just as bad as my LLMs

#38

Post by II2II »

kbenson wrote: post #30
Jensson wrote: post #29

Earlier quoted context omitted.

I have never heard of 7 being a lucky number in western culture and your link doesn't support that. 3 is a lucky number, 13 is an unlucky number, 7 is nothing to me. So I don't think its that, 7 is still a very common "random number" here even though there is no special cultural significance to it.

It's definitely used in slot machines as a lucky number. Which came first I'm not sure (but I suspect from a sibling comment in the same thread it's based on perceived commonality and primeness historically and became "lucky" in the past because of that).

While I have never heard of someone referring to 7 as a lucky number, 7 is the most common sum of two rolled dice. So I can see how people would regard it as a lucky number. Along the same lines, I assume that someone who mentions 42 as a random number has at least some interest in science fiction.

Re: People are just as bad as my LLMs

#39

Post by antihipocrat »

moffkalast wrote: post #15
tehsauce wrote: post #3

There has been some good research published on this topic of how RLHF, ie aligning to human preferences easily introduces mode collapse and bias into models. For example, with a prompt like: "Choose a random number", the base pretrained model can give relatively random answers, but after fine tuning to produce responses humans like, they become very biased towards responding with numbers like "7" or "42".

It's very funny that people hold the autoregressive nature of LLMs against them, while being far more hardline autoregressive themselves. It's just not consciously obvious.

I wonder whether we hold LLMs to a different standard because we have a long term reinforced expectation for a computer to produce an exact result?

One of my first teachers said to me that a computer won't ever output anything wrong, it will produce a result according to the instructions it was given.

LLMs do follow this principle as well, it's just that when we are assessing the quality of output we are incorrectly comparing it to the deterministic alternative, and this isn't really a valid comparison.

Post reply on HN