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

Notes on OpenAI's new o1 chain-of-thought models

simonwillison.net

Post reply on HN
Print view View original

501–510 of 659 posts

Re: Notes on OpenAI's new o1 chain-of-thought models

#501

Post by 015a »

bambax wrote: post #281

Near the end, the quote from OpenAI researcher Jason Wei seems damning to me: > Results on AIME and GPQA are really strong, but that doesn’t necessarily translate to something that a user can feel. Even as someone working in science, it’s not easy to find the slice of prompts where GPT-4o fails, o1 does well, and I can grade the answer. But when you do find such prompts, o1 feels totally magical. We all need to find…

> we all need to find harder prompts

"One of the biggest traps for engineers is optimizing a thing that shouldn't exist." (from Musk I believe)

Re: Notes on OpenAI's new o1 chain-of-thought models

#502

Post by boomchinolo78 »

nsagent wrote: post #498
MPSimmons wrote: post #106

Earlier quoted context omitted.

The failure is in how you're using it. I don't mean this as a personal attack, but more to shed light on what's happening. A lot of people use LLMs as a search engine. It makes sense - it's basically a lossy compressed database of everything its ever read, and it generates output that is statistically likely - varying degrees of likeliness depending on the temperature, as well as how many times the particular weights…

Sorry, but that does not seem to be the case. A friend of mine who runs a long context benchmark on understanding novels [1] just ran an eval and o1 seemed to improve by 2.9% over GPT-4o (the result isn't on the website yet). It's great that there is an improvement, but it isn't drastic by any stretch. Additionally, since we cannot see the raw reasoning it's basing the answers off of, it's hard to attribute this incr…

It's a good rebranding. It was getting ridiculous 3.5, 4, 4.5,

Re: Notes on OpenAI's new o1 chain-of-thought models

#503

Post by skybrian »

noisy_boy wrote: post #432
williamdclt wrote: post #164

Earlier quoted context omitted.

> Treat it as a naive but intelligent intern That’s the problem: it’s a _terrible_ intern. A good intern will ask clarifying questions, tell me “I don’t know” or “I’m not sure I did it right”. LLMs do none of that, they will take whatever you ask and give a reasonable-sounding output that might be anything between brilliant and nonsense. With an intern, I don’t need to measure how good my prompting is, we’ll usually…

I feel like it almost always starts well, given the full picture, but then for non-trivial stuff, gets stuck towards the end. The longer the conversation goes, the more wheel-spinning occurs and before you know it, you have spent an hour chasing that last-mile-connectivity. For complex questions, I now only use it to get the broad picture and once the output is good enough to be a foundation, I build the rest of it m…

Yes, I’ve seen that too. One reason it will spin its wheels is because it “prefers” patterns in transcripts and will try to continue them. If it gets something wrong several times, it picks up on the “wrong answers” pattern.

It’s better not to keep wrong answers in the transcript. Edit the question and try again, or maybe start a new chat.

Re: Notes on OpenAI's new o1 chain-of-thought models

#504

Post by ssl-3 »

TeMPOraL wrote: post #386
ssl-3 wrote: post #266

Earlier quoted context omitted.

I understand what you're saying, but the lack of acknowledgement isn't the problem I'm complaining about. The problem is the instructed lack of relevance for 99% of requests. If your sideband data included an instruction that said "This sideband data is shown to you in every request -- this means that it is not relevant to 99% of requests," then: I'd like to suggest that the for vast majority of the time, your sideba…

The "problem" is that LLMs are being asked to decide on whether, and which part of, the "sideband" data is relevant to request and act on the request in a single step. I put the "sideband" in scare quotes, because it's all in-band data. There is no way in architecture to "tag" what data is "context" and what is "request", so they do it the same way you do it with people: tell them .

Perhaps so.

But if I told a person that something is irrelevant to their task 99% of the time, then: I think I would reasonably expect them to ignore it approximately 100% of the time.

Re: Notes on OpenAI's new o1 chain-of-thought models

#505

Post by ziml77 »

fragmede wrote: post #276
croes wrote: post #270

Earlier quoted context omitted.

Intelligent? Just ask ChatGPT How many Rs are in strawberry?

https://chatgpt.com/share/66e3f9e1-2cb4-8009-83ce-090068b163... Keep up, that was last week's gotcha, with the old model.

There's randomness involved in generating responses. It can also give the wrong answer still: https://bsky.app/profile/did:plc:qc6xzgctorfsm35w6i3vdebx/po...

Re: Notes on OpenAI's new o1 chain-of-thought models

#506

Post by bluerooibos »

kristianp wrote: post #104

I tried a problem I was looking at recently, to refactor a small rust crate to use one datatype instead of an enum, to help me understand the code better. I found o1-mini made a decent attempt, but couldn't provide error free code. o1-preview was able to provide code that compiled and passed all but the test that is expected to fail, given the change I asked it to make. This is the prompt I gave: simplify this rust l…

I've been having a weird timezone issue in my Rails application that I've had a hard time getting my head around. I tried giving o1-preview the relevant code and context it needed to know and it gave answers that seemed to make sense but it still wasn't able to resolve the bug and explain exactly what was going on.

So, it seems like anything that requires some actual thought and problem-solving is tough for it to answer.

I'm sure it's just a matter of time before devs are out of work but it seems like we'll be safe for another few years anyway.

Re: Notes on OpenAI's new o1 chain-of-thought models

#507

Post by 015a »

cdelsolar wrote: post #107
elicksaur wrote: post #62

Earlier quoted context omitted.

What games have people made with ChatGPT? Do you have an example of a live, deployed game?

Yes, a gazillion of them. Someone in a scrabble Facebook group made this entirely with ChatGPT: https://aboocher.github.io/scrabble/ingpractice.html

Its actually kind of wild how obvious it is that this was not made by a human.

Re: Notes on OpenAI's new o1 chain-of-thought models

#508

Post by dev1ycan »

There's no actual improvement for real world tasks, just in-lab word prediction... it's disappointing to see so much money poured into obvious vaporware, Every 10-5 years we have a new generation of clueless VCs pouring money into something they don't understand based on lies by grifters, no different than the esports scene.

Re: Notes on OpenAI's new o1 chain-of-thought models

#509

Post by briandw »

My 12 YO and I just built a fishing game using o1 preview. Prompt: "make a top down game in pyxel. the play has to pay off a debt to a cat by catching fish. the goal is for the player to catch the giant king fish. To catch the king fish the player needs to sell the fish to the cat and get money to buy better rods, 3 levels of rod, last one can catch the king fish."

It nailed the execution. Amazing.

Re: Notes on OpenAI's new o1 chain-of-thought models

#510

Post by a_wild_dandan »

bgun wrote: post #484
deegles wrote: post #482

I was thinking about what "actual" AI would be for me and it would be something that could answer questions like "tell me every time Nicolas Cage has blinked while on camera in one of his movies". Sure, that is a contrived question, but I expect an "AI" to be capable pf obtaining every movie, watching them frame-by-frame, and getting an accurate count. All in a few seconds. Current models (any LLM) cannot do that and…

I agree. My example for something “AI” should be able to do is to create a CAD model for the Empire State Building or the Parthenon based on known facts and photos. I don’t think these are “moving the goalposts” examples, they are things that an actual intelligence capable of passing a PhD physics exam should be able to do.

I mean, I passed a physics PhD exam and I can’t model the Empire State Building. The jury is still out on whether I’m an intelligence tho.
Post reply on HN