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

Llama 2

ai.meta.com

Post reply on HN
Print view View original

251–260 of 860 posts

Re: Llama 2

#251

Post by rvnx »

onlyrealcuzzo wrote: post #237
GreedClarifies wrote: post #108

Earlier quoted context omitted.

Hard disagree. Google has made it plainly clear that they don't have anything useable in this space. Bard scores below all other commercial model. Google is getting the asses handed to them, badly. I figured that the code red would whip them into shape but the rot runs deep.

> Google has made it plainly clear that they don't have anything useable in this space. Google hasn't made their best models public because they're too expensive to run for free. > Google is getting the asses handed to them, badly. Bard has 30M active users and isn't even available in large parts of the world. They're in 2nd place - when they were pretty late to the game - that's an odd way to say someone is getting…

> Google hasn't made their best models public because they're too expensive to run for free.

?

It's the same issue with paid models.

I am paying per each request sent to Google Generative AI and this is what I get: https://i.ibb.co/4KCmz55/bard1.png

...

Re: Llama 2

#252

Post by brucethemoose2 »

What is the format for the chat models?

Alpaca and others use specific formats like:

> ### Instruction:

> ### Response:

The LLaMAv2 mentions a special chat separating token, but doesn't specify any other kind of format?

Re: Llama 2

#253

Post by Miraste »

austinkhale wrote: post #246
Miraste wrote: post #111

Earlier quoted context omitted.

This reminds me of how any day now their self driving cars are going to work right.

Their self driving cars do work? I rode in one for 30 minutes one-way on Sunday. Used it for my return trip too. No driver. Take at least 2 - 3 rides a week and have been for a few months now.

They work (most of the time) in Phoenix and SF because they've mapped every single inch of the cities by now and there are no adverse conditions. It's not scalable.

Re: Llama 2

#254

Post by californical »

barbariangrunge wrote: post #114
_b wrote: post #74

Making advanced LLMs and releasing them for free like this is wonderful for the world. It saves a huge number of folks (companies, universities & individuals) vast amount of money and engineering time. It will enable many teams to do research and make products that they otherwise wouldn't be able to. It is interesting to ponder to what extent this is just a strategic move by Meta to make more money in the end, but wh…

You have to agree to any terms they might think of in the future. Clicking download, they claim you agree to their privacy policy which they claim they can update on a whim Google's privacy policy, for example, was updated stealthfully to let them claim rights over every piece of IP you post on the internet that their crawlers can get to

You agree to their privacy policy, and they can change the privacy policy. But if you have the model and don’t interact with them, then you don’t need to agree to future revisions because you aren’t interacting with them again (unless you want newer versions)

If I buy a TV, and the store has me sign a receipt that says I agree to their privacy policy by shopping there. Then that’s fine. I don’t need to agree to any future revisions unless I go back to buy more electronics from them.

Re: Llama 2

#255

Post by andrelaszlo »

llama2.ai seems to hallucinate quite a lot.

Q: Can you list all even prime numbers less than or equal to 59?

A: Sure! The even prime numbers less than or equal to 59 are: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, 46, 48, and 50. Is there something else I can help you with?

Q: Who is [random name]?

A: [name] is a computer scientist and entrepreneur who is known for his work in the field of computer graphics and computer-aided design. He is the founder of Pixar Animation Studios and served as the company's CEO from 1986 to 2004. Under his leadership, Pixar became a leading studio for computer-animated films, producing hit movies such as "Toy Story" and "Finding Nemo." [Name] is also a pioneer in the field of digital imaging and has been recognized with numerous awards for his contributions to the tech industry.

Re: Llama 2

#256

Post by pizza »

holoduke wrote: post #189

So on a 4090 you cannot run the 70b model right?

You’d have to quantize the parameters to about 2.7 bits per parameter (24 GB / 70G * 8bits/B) - the model was likely trained at fp16 or fp32 so that would be pretty challenging. Not impossible but probably not readily available at the moment w most current quantization libraries. Quality would likely be degraded. But 2 4090s might be doable at ~4bits

Re: Llama 2

#257

Post by smoldesu »

samwillis wrote: post #233
smoldesu wrote: post #209

Earlier quoted context omitted.

If they want to own the whole stack, I don't think they have much to work with. Their highest-end server chip is a duplex laptop SOC, with maxed-out memory that doesn't even match the lowest-end Grace CPU you can buy (nevermind a fully-networked GH200). Their consumer offerings are competitive, but I don't think Apple Silicon or CoreML is ready to seriously compete with Grace and CUDA.

While Apple silicone may not be there for training, I think it's probably there for inference. I expect next years device models to launch with exclusive support for Apples own LLM based Siri.

Sure. Haswell CPUs from 2014 are "there" for inference if they have AVX support and 8gb of RAM. Inferencing isn't the problem though, not on M1 or Macbooks from 2016. Scaling a desirable (and hopefully open) GPGPU programming interface is. This is bottlenecked by both hardware and software decisions Apple has made, making a "home grown" competitive model much more unlikely in my eyes.

I agree that there is an incentive to put AI models on your OS. I just don't think Apple can own the whole stack if they want to play ball right now.

Re: Llama 2

#258

Post by PeterStuer »

jahewson wrote: post #104
chaxor wrote: post #60

Earlier quoted context omitted.

Google has far better models than llama based models. They just simply don't put them facing the public. It is pretty ridiculous that they essentially just set a marketing team with no programming experience to write Bard, but that shouldn't fool anyone into believing they don't have capable models in Google. If Deepmind were to actually provide what they have in some usable form, it would likely be quite good. Despi…

I’ve been hearing “Google has secret better models” for 7 months now. Maybe some UFOs in the hangers at Moffett Field too?

Would you believe OpenAI has vastly better models that they are not releasing publicly?

Re: Llama 2

#259

Post by messe »

jahewson wrote: post #116
whimsicalism wrote: post #4

Key detail from release: > If, on the Llama 2 version release date, the monthly active users of the products or services made available by or for Licensee, or Licensee’s affiliates, is greater than 700 million monthly active users in the preceding calendar month, you must request a license from Meta, which Meta may grant to you in its sole discretion, and you are not authorized to exercise any of the rights under thi…

Is there a good name for this kind of license? If not I propose “de-FANGed”.

How about MAANGled?

Re: Llama 2

#260

Post by villgax »

rajko_rad wrote: post #18

Hey HN, we've released tools that make it easy to test LLaMa 2 and add it to your own app! Model playground here: https://llama2.ai Hosted chat API here: https://replicate.com/a16z-infra/llama13b-v2-chat If you want to just play with the model, llama2.ai is a very easy way to do it. So far, we’ve found the performance is similar to GPT-3.5 with far fewer parameters, especially for creative tasks and interactions. Dev…

You will not use the Llama Materials or any output or results of the Llama Materials to improve any other large language model (excluding Llama 2 or derivative works thereof).

WTF kinda open for commercial model licensing is this

Post reply on HN