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

Open models by OpenAI

openai.com

Post reply on HN
Print view View original

741–750 of 909 posts

Re: Open models by OpenAI

#741

Post by blueplanet200 »

clbrmbr wrote: post #736
EagnaIonat wrote: post #734

Earlier quoted context omitted.

Nice write up! One test I do is to give a common riddle but word it slightly to see if it can actually reason. For example: "Bobs dad has five daughters, Lala, Lele, Lili, Lolo and ???" The 20B model kept picking the answer of the original riddle, even after explaining extra information to it. The original riddle is: "Janes dad has five daughters, Lala, Lele, Lili, Lolo and ???"

I don’t get it. Wouldn’t it be Lulu in both cases?

Original riddle the answer is Jane.

Re: Open models by OpenAI

#742

Post by ben_w »

cpursley wrote: post #733
ben_w wrote: post #727

Earlier quoted context omitted.

It's also extremely weird that Trump did win in 2024. If I'd been in a coma from Jan 1 2024 to today, and woke up to people saying Trump was president again, I'd think they were pulling my leg or testing my brain function to see if I'd become gullible.

You’re in a bubble. It was no surprise to folks who touch grass on the regular.

> You’re in a bubble.

Sure, all I have to go on from the other side of the Atlantic is the internet. So in that regard, kinda like the AI.

One of the big surprises from the POV of me in Jan 2024, is that I would have anticipated Trump being in prison and not even available as an option for the Republican party to select as a candidate for office, and that even if he had not gone to jail that the Republicans would not want someone who behaved as he did on Jan 6 2021.

Re: Open models by OpenAI

#743

Post by throwaway106382 »

clbrmbr wrote: post #736
EagnaIonat wrote: post #734

Earlier quoted context omitted.

Nice write up! One test I do is to give a common riddle but word it slightly to see if it can actually reason. For example: "Bobs dad has five daughters, Lala, Lele, Lili, Lolo and ???" The 20B model kept picking the answer of the original riddle, even after explaining extra information to it. The original riddle is: "Janes dad has five daughters, Lala, Lele, Lili, Lolo and ???"

I don’t get it. Wouldn’t it be Lulu in both cases?

It’s Bob or Jane.

The dad of has 5 daughters. Four are listed off. So the answer for the fifth is .

Re: Open models by OpenAI

#745

Post by BrawnyBadger53 »

clbrmbr wrote: post #736
EagnaIonat wrote: post #734

Earlier quoted context omitted.

Nice write up! One test I do is to give a common riddle but word it slightly to see if it can actually reason. For example: "Bobs dad has five daughters, Lala, Lele, Lili, Lolo and ???" The 20B model kept picking the answer of the original riddle, even after explaining extra information to it. The original riddle is: "Janes dad has five daughters, Lala, Lele, Lili, Lolo and ???"

I don’t get it. Wouldn’t it be Lulu in both cases?

Presumably Jane is a girl and therefore the fifth daughter in the original riddle.

Re: Open models by OpenAI

#746

Post by bavell »

okasaki wrote: post #716
danielvaughn wrote: post #557

Earlier quoted context omitted.

Yep that’s my biggest ask tbh. I just imagine the next Elder Scrolls taking advantage of that. Would change the gaming landscape overnight.

Games with LLM characters have been done and it turns out this is a shit idea.

There are a ton of ways to do this that haven't been tried yet.

Re: Open models by OpenAI

#747

Post by bavell »

itake wrote: post #643
captainregex wrote: post #397

Earlier quoted context omitted.

I’m still trying to understand what is the biggest group of people that uses local AI (or will)? Students who don’t want to pay but somehow have the hardware? Devs who are price conscious and want free agentic coding? Local, in my experience, can’t even pull data from an image without hallucinating (Qwen 2.5 VI in that example). Hopefully local/small models keep getting better and devices get better at running bigger…

Local micro models are both fast and cheap. We tuned small models on our data set and if the small model thinks content is a certain way, we escalate to the LLM. This gives us really good recall at really low cloud cost and latency.

I'd love to try this on my data set - what approach/tools/models did you use for fine-tuning?

Re: Open models by OpenAI

#748

Post by bavell »

JKCalhoun wrote: post #491
captainregex wrote: post #431

Earlier quoted context omitted.

love the first and am sad you’re going to be right about the second

When it was floated about that the DeepSeek model was to be banned in the U.S., I grabbed it as fast as I could. Funny how that works.

I mean, there's always torrents

Re: Open models by OpenAI

#749

Post by whynotminot »

__alexs wrote: post #691
dragonwriter wrote: post #672

Earlier quoted context omitted.

> In fact it's worse than many local models that can do it, including e.g. QwQ-32b. I'm not going to be surprised that a 20B 4/32 MoE model (3.6B parameters activated) is less capable at a particular problem category than a 32B dense model, and its quite possible for both to be SOTA, as state of the art at different scale (both parameter count and speed which scales with active resource needs) is going to have differ…

[flagged]

He’s saying there’s different goalposts at different model sizes. Is that unreasonable?

Re: Open models by OpenAI

#750

Post by bavell »

unethical_ban wrote: post #413
captainregex wrote: post #397

Earlier quoted context omitted.

I’m still trying to understand what is the biggest group of people that uses local AI (or will)? Students who don’t want to pay but somehow have the hardware? Devs who are price conscious and want free agentic coding? Local, in my experience, can’t even pull data from an image without hallucinating (Qwen 2.5 VI in that example). Hopefully local/small models keep getting better and devices get better at running bigger…

Privacy and equity. Privacy is obvious. AI is going to to be equivalent to all computing in the future. Imagine if only IBM, Apple and Microsoft ever built computers, and all anyone else ever had in the 1990s were terminals to the mainframe, forever.

Did you mean to type equality? As in, "everyone on equal footing"? Otherwise, I'm not sure how to parse your statement.
Post reply on HN