@sebsauvage j'ai l'impression aussi qu'economiquement parlant, il y a que Nvidia qui fait sont beurre et que les autres boîtes sur les investissements des banques.
There was a paper shared recently about the exponential amount of training data to get incremental performance gains in #llm#ai, but I seem to have misplaced it. Do you know what I’m referring to? Mind sharing the link if you have it?
I think one of the biggest fears people have about AI is that it isn't perfect as assumed, but that, like us humans, it takes the given information, assumes the most likely outcome, and presents it plausibly.
@anmey yeah, there’s this paradox — we kinda want computers to think like humans, but when they get plausibly good at it, we complain that they don’t think like computers anymore
I’d like to trust this story, but it fails to link to its supposed source or provide enough info to find it elsewise. A few clicks around the site makes me think that it may well be nothing but a #LLM-composed content farm. https://cosocial.ca/@kgw/112498693958537559
Llama.cpp now supports the distributed inference, meaning you can use multiple computers to speed up the response time! Network is the main bottleneck, so all machines need to be hard wired, not connected through wifi. ##LLm#AI#MLhttps://github.com/ggerganov/llama.cpp/tree/master/examples/rpc
How would anyone trust the products these people put worth? They aren’t working on making LLMs more accurate (spoiler alert: they can’t, by design), they’re working to make them more appealing to companies targeting unsuspecting consumers. By any means necessary.
The vitriol, and - honestly - ignorance around LLM-based "AI" is starting to fill my feeds from normally sane and technologically literate people.
You should be able to see through the hype and misuse. LLMs aren't encyclopedias - they're tools that are able to manipulate data of various sorts in ways that are very similar to how humans do it.
Yes, I compare LLMs to human brains. It's not the same as saying they're conscious (yet) - but the way LLMs work is apparently in many ways similar to how our brains work.
One fascinating insight into that comes from research done on what happens to the ability of LLMs to recall information as they are exposed to large and larger corpuses. Apparently they're better at recalling the early and late information, whilst starting to lose some in the middle.
In human psychology we call that the primacy and recency effect - because our brains do the same.
LLMs are absolutely awesome for a wide variety of tasks (and we have by no means found them all). Every second you spend not understanding this is a second on the way to your own irrelevance (if these tools would aid someone in your chosen area of work) or to become a grumpy old person yelling at clouds.
I just finished a productive Copilot session on a complex programming task. I came up with much of the algorithms, and wrote a lot of the code, and had to guide it a lot throughout, but credit where due, Copilot did make small but meaningful contributions along the way.
Overall, not a pair programmer but someone useful to talk to when WFH alone on complex tasks.
Enough for Copilot to earn a ✋🏽. And I like how it responded to that. It has got that part down. 😉
•This• is the compelling #LLM use case for me. If I use a translator to write messages in French I'm not forced to come up with an initial attempt and I lose the learning aspect of that.
If instead I put something into ChatGPT and it not only corrects but explains what my mistakes were that's a huge win in terms of learning from your mistakes.
(I still don't trust the thing 100% but it's also not a high stakes situation – I'm not engaging in a nuclear arms treaty after all 😅)
Absolutely unbelievable but here we are. #Slack by default using messages, files etc for building and training #LLM models, enabled by default and opting out requires a manual email from the workspace owner.
To those concerned about #slack now using your chats, including trade secrets, NDA stuff, etc., to train their #llm: #WTF did you expect using a third party with full content access to discuss those things? That they'd be gentlemen and not read your mail? That they somehow wouldn't try to find a way to monetize that juicy data? I am flabbergasted that people working for corporations just as immoral could have been that naive...
@lpwaterhouse many startups use discord for the same reason as corporates use slack
tencent has a share in discord inc. and thus an access to all the data
everybody running a startup knows that
nobody cares
idk, maybe that's applicable for startups only
@bpavuk Judging by my timeline a lot of people seem to care now; I am confused as to how they either a) didn't see it coming at all (read: naive in face of overwhelming prior examples) or b) managed to actively "not care" until it was "too late" (Hint: It was "too late" the second you put data in, not when you learned that they are using it). "a)" I can imagine, like a character in a bad novel, though it's a tall order; "b)" on the other hand... My brain does not work like that. At all. Of course there is always c) The people now clamoring are not ones that did, in fact, choose (they might have still been forced to) to use slack in the first place, but are merely expressing their indignation by stylistically putting themselves in that position to increase emotional involvement. Those are... vacuous...
The hype and utility about LLMs are overstated and will cause problems due to leadership teams in organisations (and Governments) buying into the hype.
However, they do have value as personal assistants, research assistants, and sounding boards as long as you treat all LLM output critically, especially on topics where you are not an expert.
I’m using Claude 3 Opus as a research assistant. It’s read more of the world’s info than I ever will. I am also trying out ChatGPT-4o.
@ianRobinson Among the commercial ones that are available online, Claude is definitely the best one.
Though I personally run LLMs on my own PC 90% of the time.
@hirad Agreed. I thought about running a local one (or a private Cloud LLM) and loading it with the 1000 or so articles I’ve written over the last decade. I still might.
#Anthropic is killing it with their AI game, especially for a small startup. Their models are way better than #OpenAI's, but they're focusing more on enterprise stuff rather than hyping it up. This might be a risky move since they don't have a cult following like other AI companies. Still, gotta give them props for their impressive tech. It'll be interesting to see how they balance enterprise with getting more attention from the AI community.
Looks like today I finally found a good application for #llm 's: Learning languages!
I've been attempting to learn #arabic through duolingo for a while now, without much success. I figured if there's one thing language models should be good at, it's languages. So far the thing has actually been pretty helpful.
I just posted a short tutorial on setting up a simple #LLM on your #oneplus6 running #postmarketOS . The LLM (7b alpaca in this case) is not terribly accurate but possibly useful in some cases.
Warning! You phone will get hot. I haven't tested it long enough to know if it will harm your device, but do be careful.
Oh! I have a website now! Wow! My history with websites is hit and miss, so enjoy it while it lasts. I have half a dozen posts in the works already.
@chikim The responsive interruptible conversation interface, plus the visual stuff, really makes me believe we aren't too many years off from glasses, with a voice assistant, that can describe whatever is in front of us, answer questions about it, etc. Amazing stuff!
