Anyone can direct me to alternative ways of running this on a cloud server? I want to fully host it myself on runpod or similar service. Thank you!
Meta Llama 3
361–370 of 965 posts
Re: Meta Llama 3
#362Earlier quoted context omitted.
Lex walked so that Dwarkesh could run. He runs the best AI podcast around right now, by a long shot.
I don't know Dwarkesh but I despise Lex Fridman. I don't know how a man that lacks the barest modicum of charisma has propelled himself to helming a high-profile, successful podcast. It's not like he tends to express interesting or original thoughts to make up for his paucity of presence. It's bizarre. Maybe I'll check out Dwarkesh, but even seeing him mentioned him in the same breath as Fridman gives me pause ...
Maybe you should consider that others may not share your views on Lex's lack of charisma or interesting thoughts.
Re: Meta Llama 3
#363Earlier quoted context omitted.
Upgrade to a 24GB GPU?
Any recommendations?
3090, trivially.
No reason to go 4090 as it's no more capable, and the 5090 is probably not going to have more than 24GB on it either simply because nVidia wants to maintain their margins through market segregation (and adding more VRAM to that card would obsolete their low-end enterprise AI cards that cost 6000+ dollars).
Re: Meta Llama 3
#364Earlier quoted context omitted.
Honestly, I swear to god, been working 12 hours a day with these for a year now, llama.cpp, Claude, OpenAI, Mistral, Gemini: The long context window isn't worth much and is currently creating more problems than it's worth for the bigs, with their "unlimited" use pricing models. Let's take Claude 3's web UI as an example. We build it, and go the obvious route: we simply use as much of the context as possible, given ch…
I don't need a million tokens, but 8k is absolutely too few for many of the use cases that I find important. YMMV.
I don't think it's a YMMV thing: no one claims it is useless, in fact, there's several specific examples of it being necessary.
Re: Meta Llama 3
#365Earlier quoted context omitted.
Lex walked so that Dwarkesh could run. He runs the best AI podcast around right now, by a long shot.
I don't know Dwarkesh but I despise Lex Fridman. I don't know how a man that lacks the barest modicum of charisma has propelled himself to helming a high-profile, successful podcast. It's not like he tends to express interesting or original thoughts to make up for his paucity of presence. It's bizarre. Maybe I'll check out Dwarkesh, but even seeing him mentioned him in the same breath as Fridman gives me pause ...
I mostly agree with you. I listened to Fridman primarily because of the high profile AI/tech people he got to interview. Even though Lex was a terrible interviewer, his guests were amazing.
Dwarkesh has recently reached the level where he's also interviewing these high profile AI/tech people, but it's so much more enjoyable to listen to, because he is such a better interviewer and skips all the nonsense questions about "what is love?" or getting into politics.
Re: Meta Llama 3
#366Awesome, but I am surprised by the constrained context window as it balloons everywhere else. Am I missing something? 8k seems quite low in current landscape.
Based on your use cases.
I thought it's not hard to push the window to 32K or even 100k if we change the position embedding
Re: Meta Llama 3
#367Is there a download link for this model like LLAMA2 or is it going to be exclusively owned and operated by Meta this time?
https://huggingface.co/meta-llama/Meta-Llama-3-8B https://huggingface.co/meta-llama/Meta-Llama-3-70B https://llama.meta.com/llama-downloads https://github.com/meta-llama/llama3/blob/main/download.sh
Thank you kind stranger
Re: Meta Llama 3
#368Earlier quoted context omitted.
Having an engineering mindset is not the same as never making mistakes (or never being too early to the market). The only way you won’t make those mistakes and keep a perfect record is if you never do anything major or step out of the comfort zone. If Apple didn’t try and fail with Newton[0] (which was too early to the market for many reasons, both tech-related and not), we might’ve not had iPhone today. The engineer…
His engineering mindset made him blind to the fact the metaverse was a product that nobody wanted or needed. In one of the Fridman interviews, he goes on and on about all the cool technical challenges involved in making the metaverse work. But when Fridman asked him what he likes to do in his spare time, it was all things that you could precisely not do in the metaverse. It was baffling to me that he failed to connec…
I don't think that was the issue. VRChat was basically the same idea but done in a more appealing way and it was (still is) wildly popular.
Re: Meta Llama 3
#369Earlier quoted context omitted.
That's coz he is a founder CEO. Those guys are built different. It's rare for the careerist MBA types to match their passion or sincerity. There are many things I can criticize Zuck for but lack of sincerity for the mission is not one of them.
It is just the reverse: he is successful because he is like that and lots of founder ceos are jellies in comparison
I dunno. I find a conviction in passion in founder CEOs that is missing in folks who replace them.
Compare Larry & Sergey with Pichai, or Gates with Balmer.
Re: Meta Llama 3
#370They've got a console for it as well, https://www.meta.ai/ And announcing a lot of integration across the Meta product suite, https://about.fb.com/news/2024/04/meta-ai-assistant-built-wi... Neglected to include comparisons against GPT-4-Turbo or Claude Opus, so I guess it's far from being a frontier model. We'll see how it fares in the LLM Arena.
[flagged]
I haven't tried Llama 3 yet, but Llama 2 is indeed extremely "safe." (I'm old enough to remember when AI safety was about not having AI take over the world and kill all humans, not when it might offend a Puritan's sexual sensibilities or hurt somebody's feelings, so I hate using the word "safe" for it, but I can't think of a better word that others would understand).
It's not quite as bad as Gemini, but in the same class where it's almost not useful because so often it refuses to do anything except lecture. Still very grateful for it, but I suspect the most useful model hasn't happened yet.