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

401–410 of 860 posts

Re: Llama 2

#401

Post by brucethemoose2 »

DebtDeflation wrote: post #218
swyx wrote: post #177

Earlier quoted context omitted.

if you red team the 13b and the 70b and they pass, what is the danger of 34B being significantly more dangerous? edit: turns out I should RTFP. there was a ~2x spike in safety violations for 34B https://twitter.com/yacineMTB/status/1681358362057883680?s=2...

A 34B model is probably about the largest you can run on a consumer GPU with 24GB VRAM. 70B will require A100's or a cloud host. 13B models are everywhere already. I'm sure this was a very deliberate choice - let people play with the 13B model locally to whet their appetite and then they can pay to run the 70B model on Azure.

70B should work reasonably well on 64GB CPU RAM + any decent GPU, or maybe a 24GB GPU + 32GB CPU RAM.

Re: Llama 2

#402

Post by echelon »

brucethemoose2 wrote: post #176
andy99 wrote: post #20

Another non-open source license. Getting better but don't let anyone tell you this is open source. http://marble.onl/posts/software-licenses-masquerading-as-op...

Is a truly open source 2 trillion token model even possible? Even if Meta released this under Apache 2.0, there's the sticky question of the training data licenses.

> Even if Meta released this under Apache 2.0, there's the sticky question of the training data licenses.

You need to pay Disney royalties out of every paycheck because you watched Toy Story and Star Wars. These movies updated your own neural weights.

Re: Llama 2

#403

Post by nottheengineer »

arbuge wrote: post #154
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…

I like the way the playground allows easy modification of the system prompt. I suggest adding "You are very cranky." to the default prompt for interesting results.

Holy shit, I've never seen an AI go schizophrenic this hard.

That's my first try: https://gist.github.com/miguel7501/983f794e13cc762eb6274c9b2...

Re: Llama 2

#404

Post by pmarreck »

I've actually encountered situations with the current gen of "curated" LLM's where legitimate good-actor questions (such as questions around sex or less-orthodox relationship styles or wanting a sarcastic character response style, etc.) were basically "nanny-torpedoed", if you know what I mean.

To that end, what's the current story with regards to "bare" open-source LLM's that do not have "wholesome bias" baked into them?

Re: Llama 2

#405

Post by brucethemoose2 »

LoganDark wrote: post #322

I just tested the 13b-chat model and it's really good at chatting, even roleplaying, seemingly much better than other models I've tried (including uncensored ones like Pygmalion), fun!! It also doesn't seem to get constantly tripped up by second-person :D

Pygmalion 13B was kind if a dud.

Have you tried Chronos-Hermes 13B? Thats SOTA 13b roleplaying, as far as I know.

Re: Llama 2

#406

Post by stOneskull »

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…

> say the tracklist for lou reed album 'new york'

>Sure! The tracklist for Lou Reed's album "New York" is as follows:

>"Heroin" >"New York City" >"Kicks" >"Betrayed" >"I'm Waiting for the Man" >"The Kids" >"Oh! Sweet Honey Pie" >"Sunday Morning" >"Romeo Had Juliette" >"The Bed"

>I hope that helps! Let me know if you have any other questions.

Re: Llama 2

#407

Post by pmarreck »

CSMastermind wrote: post #332
jug wrote: post #275

Earlier quoted context omitted.

Still fails my hippo test! > Yes, hippos are excellent swimmers. They spend most of their time in the water, where they feed on aquatic plants and escape the heat of the savannah. In fact, hippos are one of the best swimmers among all land mammals. But that's fine. Most do. Hippos don't swim. They walk or hop/skip at best underwater.

There's a few prompts that I use with every model to compare them. One of the simplest ones is: > When does the bowl of the winds get used in the wheel of time books? LLaMA2 fails pretty hard: > The Bowl of the Winds is a significant artifact in the Wheel of Time series by Robert Jordan. It is first introduced in the third book, "The Dragon Reborn," and plays a crucial role in the series throughout the rest of the bo…

> Please write a function in JavaScript that takes in a string as input and returns true if it contains a valid roman numeral and false otherwise.

Your question actually isn't worded concisely enough. You don't specify whether the string can merely contain the roman numeral (plus other, non-roman-numeral text), or must entirely consist of just the roman numeral. The way "if it contains" is used colloquially, could imply either.

I'd use either "if it IS a roman numeral" if it must consist only of a roman numeral, and "if there exists a roman numeral as part of the string" or some such, otherwise.

Re: Llama 2

#408

Post by marricks »

_2d30 wrote: post #291
rileyphone wrote: post #279

Earlier quoted context omitted.

Wow, that looks so bad from an anti-trust/competitiveness standpoint. M$ is embracing AI just like it embraced the internet 25 years ago.

Yeah and look how they extended and extinguished that!

I mean, they dominated internet browsers by being the default option until they sucked at it so hard people downloaded alternatives.

I’m not sure you want to invite the comparison.

Re: Llama 2

#409

Post by brucethemoose2 »

appleflaxen wrote: post #183

When you ask about the harmful effect of Facebook on society: > I understand that you have concerns about Facebook and their impact on society. However, it's important to note that Facebook is a complex organization with many different teams and initiatives, and it's not accurate to make sweeping generalizations about the company as a whole. While some decisions and products may be controversial or have unintended co…

I would love to see their "Meta alignment" dataset. They must have one, as LLaMA recognizes stuff about itself.

Re: Llama 2

#410

Post by BryanLegend »

cedarseagull wrote: post #387
_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…

I think it's a defense against anti-trust attacks. Bell Labs did a TON of this in 60's and 70's when they owned most of the phone lines, service contracts, AND TELEPHONES THEMSELVES in the USA. I believe companies open source their research efforts as a thing to point to when regulators come so they can say "look at all the good our cash-faucet monopoly provides the economy!!"

Facebook can surely use the reputation points
Post reply on HN