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

Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

cerebras.net

Post reply on HN
Print view View original

171–180 of 231 posts

Re: Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

#171

Post by fnordpiglet »

kweingar wrote: post #166
fnordpiglet wrote: post #115

Earlier quoted context omitted.

It’s because they’re confused in thinking human intelligence isn’t learned stochastic expectation.

This seems backward to me. Wouldn’t you be less impressed by ChatGPT if you thought that thought that human intelligence worked the same way as LLMs? If humans have some special sauce different from the computer, then it’s crazy that ChatGPT can emulate human writing so well. If humans are also just statistical models, then of course you can throw a big training set at some GPUs and it’ll do the same thing. Why shoul…

Well, I don’t they work exactly like the human mind, because we are likely a more complex ensemble model. But a) it’s impressive because it’s reaching a goal of computing for generations and as a computer scientist I’m excited, b) it’s corpus of learnings is the amazing collection of everything that we built with the modern internet, so it’s breadth and depth is astounding. The act of creating what we have naturally is itself exciting though.

Re: Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

#173

Post by dannyobrien »

jazzkingrt wrote: post #34
2bitencryption wrote: post #26

This type of article (or press release, or whatever you want to call it) is exactly what makes the future so interesting. The cat is out of the bag, the genie is out of the bottle, the confetti has left the cannon[0]. It's tempting to see a world dominated by Google Bard, ChatGPT, Bing Search, etc. And no doubt, they will be huge players, with services that are far more powerful than anything that can be run on the e…

Serious question: is it typical to describe client-side computing as "on the edge"? I thought running something on the edge referred to running it in close network proximity to the user, rather than users having control and running things themselves.

I've used "edge" in this context for around 15 years[1], and I've always intended it to mean "at the edge of the network", which can include being on the other side of the world to a user.

[1] from https://www.oblomovka.com/wp/2007/08/ at least

Re: Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

#174

Post by sitkack »

worldsayshi wrote: post #169
worldsayshi wrote: post #149

Earlier quoted context omitted.

I think the thrilling part is that it's a somewhat atomic concept that can somewhat convincingly be proven to not exist in the training data. While poetry is more impressive if it's as original it's harder to show that it's not just stitched together from the training data.

I just asked GPT-4 to come up with more such "provably original idioms": """ Here are a few more examples of idioms with meaningful and provable atomic originality: "The kite has touched the stars" - This phrase could mean that someone has achieved a seemingly impossible goal or reached a level of success that was thought to be unattainable. "The paint has mingled on the canvas" - This idiom might convey the idea tha…

Yeah but can it beatbox?

Re: Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

#175

Post by bloaf »

wing-_-nuts wrote: post #161
2bitencryption wrote: post #26

This type of article (or press release, or whatever you want to call it) is exactly what makes the future so interesting. The cat is out of the bag, the genie is out of the bottle, the confetti has left the cannon[0]. It's tempting to see a world dominated by Google Bard, ChatGPT, Bing Search, etc. And no doubt, they will be huge players, with services that are far more powerful than anything that can be run on the e…

>So every time you're amazed by something chat-gpt4 says, remember that soon this will be in your pocket. I want to believe you, but I'm ignorant of the hardware requirements for these things. How soon do you think we'd be able to run something reasonably gpt4-like on, say, a 4090?

It's literally already here: https://twitter.com/thiteanish/status/1635678053853536256

Re: Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

#177

Post by alfor »

akiselev wrote: post #68
lioeters wrote: post #50

Earlier quoted context omitted.

Yes, yes, and yes. I'm waiting for an actually open AI that can run on the edge, purely on commodity hardware like our laptops and phones - it's inevitable. I imagine this "cat out of the bag" situation, the democratization and commodification of powerful technology accessible and affordable to the public, is similar to what's happening with single-board computers and microcontrollers like Raspberry Pi, Arduino, ESP3…

After using ChatGPT 4 extensively for a few days, I think we're probably only a few years away from the first generation of truly conversational assistants ala Jarvis in Iron Man. Between LangChain and existing voice recognition software, we've already 95% of the way there, it just needs to be packaged up into a UI/UX that makes sense. These local models are absolutely critical for that to happen though. I'm hitting…

GPT-4 training was finished last summer. Karpathy is working on a JARVIS there. I think they already have something quite good for internal use, maybe trying to make it "safe" before releasing it.

My guess: we will have a competent JARVIS(minus the holograms) this year.

Re: Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

#178

Post by OJFord »

kweingar wrote: post #166
fnordpiglet wrote: post #115

Earlier quoted context omitted.

It’s because they’re confused in thinking human intelligence isn’t learned stochastic expectation.

This seems backward to me. Wouldn’t you be less impressed by ChatGPT if you thought that thought that human intelligence worked the same way as LLMs? If humans have some special sauce different from the computer, then it’s crazy that ChatGPT can emulate human writing so well. If humans are also just statistical models, then of course you can throw a big training set at some GPUs and it’ll do the same thing. Why shoul…

I suppose it depends on the extent to which you're impressed by the concept of LLMs, I can see it both ways.

To explain the other way to your thinking: human intelligence is the same; holy crap they cracked robotic 'human' intelligence, it works exactly the same way.

Re: Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

#179

Post by ericd »

dukeofdoom wrote: post #162

I wonder how decrotive our world will become as a consequence of how cheap it will become to make art using AI. I kind of want 3d marble statues and baroque art of a future reinasance everywhere. But wonder if we will turn minimalistic as a response.

I've been wondering about the best way to print large format versions of custom Renaissance-style paintings with goofy subjects for our walls at home. I guess I have to figure out how to best upscale the output first.

Re: Cerebras-GPT: A Family of Open, Compute-Efficient, Large Language Models

#180

Post by anon291 »

IshKebab wrote: post #159
brucethemoose2 wrote: post #18

FYI: Cerebras's nodes are very different than your typical Nvidia training nodes: https://www.anandtech.com/show/16626/cerebras-unveils-wafer-... Each individual "chip" has 40GB of SRAM vs ~76MB for the Nvidia H100, and networked pools of external RAM, SSDs and such. Thats why the training architecture is so different.

It's a pretty mad architecture tbh. Compile times must be absolutely insane. Also Tesla's Dojo also uses a manufacturing technique that has basically obsoleted their WSI design already.

Compile times are not a whole lot different than any other large model build. It's a kernel based compilation pipeline and the kernels are simply tiled over a 'core' area in the weight streaming architecture.
Post reply on HN