ilaksh

40 ms·

781.
by ilaksh 2y ago
I wasn't trying to connect all those ideas like that. I was just trying to bring up concepts like artificial "offspring" etc. because I think they are relevant to the discussion and interesting. I think you can look at the po
782.
by ilaksh 2y ago
Artificial wombs are fascinating and I hope they eventually become a viable alternative for mothers. But I see them as being perhaps a somewhat more distant future possibility to fully replace a real womb due to the all of challenges. When
783.
by ilaksh 2y ago
I don't know about now, but I think this is going to be another solid use case for robots, once we get to a certain high level of dexterity and capability in general. Could be 5-10 years or more. Although an easier solution would be fo
784.
by ilaksh 2y ago
Another fun read from Marshall Brain: "The Second Intelligent Species: How Humans Will Become as Irrelevant as Cockroaches" available on Amazon.
785.
by ilaksh 2y ago
Your post would make more sense to me if you were specific about the models. It's like if you were asking about how to get reliable transportation from a car and didn't specify which model of cars you were considering. o1-preview
786.
by ilaksh 2y ago
That's correct. The whole thing is being promoted in a deliberately misleading way by multiple groups to get clicks.
787.
by ilaksh 2y ago
It just ran 'sudo'.
788.
by ilaksh 2y ago
His system instructions include this: "In general, if there's a way to continue without user assistance, just continue rather than asking the user something. Always include a bash command in your message unless you need to wait fo
789.
by ilaksh 2y ago
The models are very powerful. This can help anyone, including scammers. The number of scams will be enormous. I just hope people are also able to see how useful it is for non-scumbags and don't let the scammers ruin it for everyone els
790.
by ilaksh 2y ago
Prompt enhancement is now a standard feature in many image generation tools.
791.
by ilaksh 2y ago
Pretty smart model. Here's one I made: https://replicate.com/p/6ez0x8xqvsrga0cjadg8m7bah0
792.
by ilaksh 2y ago
They have evidence that inference time and inference time during training can continue to increase the reasoning abilities. They do not actually need any further technology development to continue to add profitable products. There are numer
793.
by ilaksh 2y ago
You're right, having a voice conversation for any reason is just so passe these days. They should stop adding microphones to phones and everything. So old-fashioned and inefficient. And who wants to ever have to actually talk to someon
794.
by ilaksh 2y ago
I think it's just not a super smart model. They had to make a slight compromise to keep the latency low. The naturalness of the conversation that they did achieve is a great technical accomplishment with these types of constraints thou
795.
by ilaksh 2y ago
Oh sorry I didn't see that. Got it. $0.24 per minute.
796.
by ilaksh 2y ago
It is expensive. They charge in 6 second increments. I have not found anywhere that says how much per 6 second stream. Okay found it, $0.24 per minute, on the bottom of the pricing page. That means they can spend $14/hour on GPU and st
797.
by ilaksh 2y ago
This is so amazing. What's the base rate for streaming with the API? Can you add that to the Pricing page please?
798.
by ilaksh 2y ago
Nice. There is also 'exa'.
799.
by ilaksh 2y ago
It would be ideal if they made the SoundStorm model available via API.
800.
by ilaksh 2y ago
I guess I am a little cynical, but one alternative interpretation of this article is that five year olds can be quite difficult, and it is very tempting to just waste 4 hours to avoid dealing with a temper tantrum.
801.
by ilaksh 2y ago
I have seen at least one experiment running a language model or other neural network on (small scale) memory-based computing substrates. That suggests less than 1-2 years to apply them immediately to existing tasks once they are scaled up i
802.
by ilaksh 2y ago
I think it's just ignorance and timidity on the part of investors. Memristor or memory-computing startups are surely the next trend in investing within a few years. I don't think it's necessarily demand or any particular calc
803.
by ilaksh 2y ago
So you are basically saying "it failed on some of my Rust tasks, and those other languages aren't even real programming languages, so it's useless". I've used LLMs to generate quite a lot of Rust code. It can defini
804.
by ilaksh 2y ago
Which model exactly? You understand that every few months we are getting dramatically better models? Did you try the one that came out within the last week or so (o1-preview).
805.
by ilaksh 2y ago
Isn't this already the case for large portions of GPUs? Like, many of the blocks would be systolic arrays? I think the next step is arrays of memory-based compute.
806.
by ilaksh 2y ago
How far are we from memory-based computing going from research into competitive products? I get the impression that we are already well passed the point where it makes sense to invest very aggressively to scale up experiments with things li
807.
by ilaksh 2y ago
Serious question: what did Steve Jobs actually figure out while on a long walk? Did Jobs actually invent any of the user interface or user experience concepts or marketing? Is the idea of making a high quality or perceived high quality prod
808.
by ilaksh 2y ago
Claude 3.5 Sonnet Project / Artifacts
809.
by ilaksh 2y ago
"The truth is that Watsonx [IBM’s generative AI offering] isn’t even available to employees to attempt to try and help automate some meaningless task. It's so far behind OpenAI and ChatGPT that it’s not even close."
810.
by ilaksh 2y ago
Yes it was called Watson, but I don't think the technology is related to what they used. So maybe not an LLM at that time. "Watson" just became a brand name for IBM.

More ›