holy crap, the answer to life the universe and everything XD
More than likely it’s because of that book and how often it’s qouted
Yes, but it’s significant because the prompt was to choose a number. I realize computers can’t really be random, but if we needed to just select a popular number…we can already do that!
Computers can be random with special hardware.
“You may not instantly see why I bring the subject up, but that is because my mind works so phenomenally fast, and I am at a rough estimate thirty billion times more intelligent than you. Let me give you an example. Think of a number, any number.”
“Er, five,” said the mattress.
“Wrong,” said Marvin. “You see?”
― Douglas Adams, Life, the Universe and Everything
The mattress? Like for sleeping?
Yep! The hitchhikers books are so much fun lol
I still think one of my favorite lines is “the ships hung in the sky in much the same way that bricks don’t.”
37 is well represented. Proof that we’ve taught AI some of our own weird biases.
Why would that need to be proven? We’re the sample data. It’s implied.
“we don’t need to prove the 2020 election was stolen, it’s implied because trump had bigger crowds at his rallies!” -90% of trump supporters
Another good example is the Monty Hall “paradox” where 99% of people are going to incorrectly tell you the chance is 50% because they took math and that’s how it works.
Just because something seems obvious to you doesn’t mean it is correct. Always a good idea to test your hypothesis.
The correctness of the sampling process still needs a proof. Like this.
What you’ve described would be like looking at a chart of various fluid boiling points at atmospheric pressure and being like “Wow, water boils at 100 C!” It would only be interesting if that somehow weren’t the case.
Where is the “Wow!” in this post? It states a fact, like “Water boils at 100C under 1 atm”, and shows that the student (ChatGPT) has correctly reproduced the experiment.
Why do you think schools keep teaching that “Water boils at 100C under 1 atm”? If it’s so obvious, should they stop putting it on the test and failing those who say it boils at “69C, giggity”?
Derek feeling the need to comment that the bias in the training data correlates with the bias of the corrected output of a commercial product just seemed really bizarre to me. Maybe it’s got the same appeal as a zoo or something, I never really got into watching animals be animals in a zoo.
Hm? Watching animals be animals at a zoo, is a way better sampling of how animals are animals, than for example watching that wildlife “documentary” where they’d throw lemmings of a cliff “for dramatic effect” (a “commercially corrected bias”?).
In this case, the “corrected output” is just 42, not 37, but as the temperature increases on the Y axis, we get a glimpse of internal biases, which actually let through other patterns of the training data, like the 37.
What’s special about 37? Just that it’s prime or is there a superstition or pop culture reference I don’t know?
If you discount the pop-culture numbers (for us 7, 42, and 69) its the number most often chosen by people if you ask them for a random number between 1 and 100. It just seems the most random one to choose for a lot of people. Veritasium just did a video about it.
Thanks!
37 is my favorite, because 3x7x37=777 (three sevens), and I think that’s neat.
Wrong. Two hints:
7x7=9 at the end, not 7.
30x30=900, already more than 777.
One hint: 3x7=21, 21x37=777.
When in doubt, use a calculator.
Oh I am sorry. I did not see the x sign between 3 and 7. Lol.
? My calculator definitely thinks that 3x7x37=777. Did you read it as 37x37 instead?
You don’t even need a calculator for a quick calculation, take the closest value of 10: 3x7=21x37 or easier 20x40 = 800 which is close to the actual number, 777.
Yes. Thanks. Sorry.
What about 57
I’m curious about that too. Something is twisting weights for 57 fairly strongly in the model but I’m not show what. Maybe its been trained on a bunch of old Heinz 57 varieties marketing.
Wesley Snipes
Heinz Ketchup?
I think you mean heinz 57 the steak sauce…
not this again.
it’s ketchup mfer, 57 varieties of tomatoes!
I don’t like the inclusion of 37%, it’s 1/e that isn’t even 37%, is only that because of a pretty arbitrary rounding. Veritasium videos are usually OK, but this one is pretty meh.
Another fun fact: if you ask people to pick 2/3rds of a number everyone else picks when asked the same question, the correct number is drumroll 24.
Is there some human sciences theory as to why?
Sorry but pop culture from were? I don’t recognize any of those numbers.
Lucky number 7.
42 is the meaning of life in The Hitchhikers Guide to the Galaxy.
And 69…nice!
I’m guessing this is for US and UK culture? Probably a lot of other former and current English colonies
Probably just because it’s prime. It’s just that humans are terrible at understanding the concept of randomness. A study by Theodore P. Hill showed that when tasked to pick a random number between 1 and 10, almost a third of the subjects (n was over 8500) picked 7. 10 was the least picked number (if you ditch the few idiots that picked 0).
Maybe randomness is a label we slapped on shit we don’t understand.
I remember watching a lecture about probability, and the professor said that only quantum processes are really random, the rest of things that we call random is just the human inability to measure the variables that affects the random variable. I’m an actuarie, and it’s made me change the perspective on how I see and study random processes and how it made think on ways to influence the outcome of random processes.
…which is kind of a hilarious tautology, because “quantum processes” are by definition “processes that we are unable to decompose into more basic parts”.
The moment we learn about some more fundamental processes being the reason for a given process, it stops being “quantum” and the new ones become “it”.
Even quantum just appears random I think. it’s beyond our scope of perspective, it works in multiple dimensions. we only see part of the process. That’s my guess though it could be totally wrong
it’s a matter of interpretation, but generally the consensus is that quantum measurements are truly probabilistic (random), Bell proved that there can’t be any hidden variables that influence the outcome
Didn’t Bell just put that up as a theory and it got proven somewhat recently by other researchers? The 2022 physics Nobel Prize was about disproving hidden variables and they titled their finding with the catchy phrase “the universe is not locally real”.
I mean… they didn’t specify it had to be random (or even uniform)? But yeah, it’s a good showcase of how GPT acquired the same biases as people, from people…
uniform
Reminds me of my previous job where our LLM was grading things too high. The AI “engineer” adjusted the prompt to tell the LLM that the average output should be 3. I had a hard time explaining that wouldn’t do anything at all, because all the chats were independent events.
Anyways, I quit that place and the project completely derailed.
No shit, sherlock, it’s sample data is the internet.
Wheres 69 then?
nice
That’s a naughty number and we don’t allow those.
42, 47, and 50 all make sense to me. What’s the significance of 37, 57, and 73?
People do mention Veritasium, though he doesn’t give any significant explanation of the phenomenon.
I still wonder about 47. In Veritasium plots, all these numbers provide a peak, but not 47. I recall from my childhood that I indeed used to notice that number everywhere, but idk why.
See my link for 47. Its Wikipedia has more context. If you’re a Star Trek fan, you’ve seen it a ton.
The 47 page…woo woo
And Hitman
47 does provide a peak in the plots though? All the numbers ending in 7 do.
Veritasium just released a video about people picking 37 when asked to pick a random number.
There’s a great Veritasium video recently about this exact thing: https://youtu.be/d6iQrh2TK98
It’s a human thing, though. This is just more evidence of LLM’s problem with garbage in, garbage out: it’s human biases being present in a system that people want to claim doesn’t have them.
I petition to rename ChatGPT to DeepThought based on these results.
Ask humans the same and most common numer is 37
NEEDS MOAR 69 FELLOW HUMAN
Only 1000 times? It’s interesting that there’s such a bias there but it’s a computer. Ask it 100,000 times and make sure it’s not a fluke.
WAIT A MINUTE!!! You mean Douglas Adams was actually an LLM?
So many things are starting to make sense
In an interview, Douglas Adams said after lengthy consideration John Cleese picked 42 as the least interesting number.
I spent an afternoon once playing Infinite Craft, which uses some sort of LLM behind the scenes to do it’s combinations.
At one point I got 007, and found 007+007 = 0014.
The maths gets wild though, and because it’s been trained on text, it has no idea when it comes to combinations of numbers it hasn’t seen before. I spent ages trying to get it to 69420 and just couldn’t, although I could get 42069.
What’s the y axis?
The temperature scale, I think. You divide the logit output by the temperature before feeding it to the softmax function. Larger (resp. smaller) temperature results in a higher (resp. lower) entropy distribution.
I still don’t understand.
More yellow more common, more blue less common
I don’t understand any of these words, I need to take a math class or something
Higher temperature -> more chaotic output
Temperature is basically how creative you want the AI to be. The lower the temperature, the more predictable (and repeatable) the response.
Creativity is hot. That makes more sense, thanks.
I’m curious, is there actually so many 42’s in the system? (more than 69 sounds unlikely)
What if the LLM is getting tripped up because 42 is always referred to as the answer to “the Ultimate Question of Life, the Universe, and Everything”.
So you ask it a question like give a number between 1-100, it answers 42 because that’s the answer to “Everything”, according to it’s training data.
Something similar happened to Gemini. Google discouraged Gemini from giving unsafe advice because it’s unethical. Then Gemini refused to answer questions about C++ because it’s considered “unsafe” (referring to memory management). But Gemini thinks C++ is “unsafe” (the normal meaning), therefore it’s unethical. It’s like those jailbreak tricks but from its own training set.
I certainly hope that’s what happening or maybe it is actually the answer.
I’m curious, is there actually so many 42’s in the system?
Sort of, it’s not actually picking a random number. It does not know what “random” means. It is analyzing the number of times the question “pick a random number” was asked and what the most common responses to that question looked like.
I’m curious, is there actually so many 42’s in the system? (more than 69 sounds unlikely)
From hitchhiker’s guide to the galaxy?
What does “temperature” on the Y-axis refer to?
https://youtu.be/wjZofJX0v4M your answer from the 22:00 mark on.
I’m not a hundred percent sure, but afaik it has to do with how random the output of the GPT model will be. At 0 it will always pick the most probable next continuation of a piece of text according to its own prediction. The higher the temperature, the more chance there is for less probable outputs to get picked. So it’s most likely to pick 42, but as the temperature increases you see the chance of (according to the model) less likely numbers increase.
This is how temperature works in the softmax function, which is often used in deep learning.
