smeeth

7 ms·

31.
by smeeth 1y ago
Their FOSS local inference service didn't go anywhere. This isn't Anaconda, they didn't do a bait and switch to screw their core users. It isn't sinful for devs to try and earn a living.
32.
by smeeth 1y ago
You don't need the analogy. If you have a tool that does a job for you your capacity to do the job degrades alongside other associated skills. Tools that do many things and tools that do a small number of things are still tools. > &
33.
by smeeth 1y ago
It's worth noting this is the exact argument people used against adopting electric calculators.
34.
by smeeth 1y ago
The main limitation of tokenization is actually logical operations, including arithmetic. IIRC most of the poor performance of LLMs for math problems can be attributed to some very strange things that happen when you do math with tokens. I&
35.
by smeeth 1y ago
+1, this is the exact reason I started using uv. Extremely convenient. For some reason uv pip has been very slow, however. Unsure why, might be my org doing weird network stuff.
36.
by smeeth 1y ago
I suspect its something to do with the following: When humans get stuck solving problems they often go out to acquire new information so they can better address the barrier they encountered. This is hard to replicate in a training environme
37.
by smeeth 1y ago
They have a deal with Cerebras for inference. https://www.cerebras.ai/blog/mistral-le-chat
38.
by smeeth 1y ago
It did buy them a hell of a lot of marketing, Telegram has a billion users and those users are mostly in countries where Musk's brand isn't as toxic.
39.
by smeeth 1y ago
I don't like Grok that much, but there's nothing particularly new or interesting about this deal. Telegram has a big international audience, it makes sense someone would pay to be the default for user adoption reasons. "Feels
40.
by smeeth 1y ago
I hate hate hate putting Deep Think behind a paywall. It's an ease-of-use tax. I fully expect to be able to get it over API through Poe or similar for way cheaper. Just have usage limit tiers!
41.
by smeeth 1y ago
You wrote "culture." If you meant "politics," then write that. I'm not capable of reading your mind. Accusing me of projecting when you invoked "culture" to explain poverty in a state like Mississippi is h
42.
by smeeth 1y ago
It's truly remarkable how much compassion liberals purport to have for the poor before, apparently without any self-awareness, saying something like this. > This article is advocating that using poverty-adjusted scores is a better w
43.
by smeeth 1y ago
They would do this if their mission was what you wish it was. But it isn't, so they won't.
44.
by smeeth 1y ago
Not at all, even if there was snark intended I think its a fair question. I resisted ranting about it (I hope you'll indulge me now), but in general I don't much like the economics commentariat. As best I can tell, "respectab
45.
by smeeth 1y ago
No (likely because American). It's entirely possible my interpretation of his opinion of GDP is totally off, but that's what I gleaned from this single article. If I'm wrong, it's at least a bit his fault. Do you have an
46.
by smeeth 1y ago
I'm not going to argue that accounting is all kinds of messed up (totally is), but your comment isn't factually accurate. > This is true, but it used to measure (or at least try to account for) decay in private companies. No, t
47.
by smeeth 1y ago
Elon is definitely wrong. The author seems to be advocating running the economy on vibes, however. > The real problem with GDP ... is that it tempts us to view all human progress as an amorphous score, which we can start maximising just
48.
by smeeth 1y ago
When we talk about harm on the internet, we often separate the activity from the harm, e.g., a problem with [activity] is that it could cause [harm]. But it's now clear that the activity is the harm. The issue isn’t that using TikTok m
49.
by smeeth 1y ago
In fairness, 4o was like this until very recently. I suspect it comes from training on COT data from larger models.
50.
by smeeth 2y ago
It’s not just that people disagree on whether or not to do these things, it’s also that they disagree on what helps people/the world. An evangelical and an atheist will probably disagree about the helpfulness of spreading the gospel, f
51.
by smeeth 2y ago
I could be totally wrong (please correct me if so), but IMO there are two things happening that weren't totally discussed. 1) My understanding is that most the "researchers" and even students at ATI have a main role at a home
52.
by smeeth 2y ago
I'm not a doctor, but my understanding is that lower blood pressure can indeed lead to a higher heart rate under certain circumstances. For instance, people taking minoxidil for hair loss can experience lower blood pressure, which lead
53.
by smeeth 2y ago
Related startup: Reflect Orbital plans to put mirrors in space so that light can be reflected back onto ground-based panels at night. A base assumption is often that orbital solar requires a panel in space, but even that might not be the ca
54.
by smeeth 2y ago
I know, my point is that I fail to see how the FTX collapse would lead them to a tone change. > Maybe FTX did teach them something The new tone is all about business fundamentals and the FTX collapse had literally nothing to do with busi
55.
by smeeth 2y ago
I don’t think FTX is the right contrast here. This document is mostly about business realism: revenue, margins, etc. VCs did get burned by speculative investing in the 2019-21 period but FTX wasn’t like the others. At the time of its collap
56.
by smeeth 2y ago
I wonder sometimes if the reason the US doesn't have universal healthcare is the 1987 Emergency Medical Treatment and Active Labor Act (EMTALA), which requires hospitals to provide life-saving emergency care regardless of a patient&#x2
57.
by smeeth 2y ago
So I totally get that people are upset, but it’s probably a good idea to remember that SO was on the path towards going out of business. If they didn’t change something they would simply cease to exist. Does anyone have a better idea?
58.
by smeeth 3y ago
I had an experience like this once! My my laptop would inexplicably and intermittently stop connecting to the internet. It turned out my bluetooth headset was using the same band as the wifi but I only figured this out after a few months an
59.
by smeeth 3y ago
Check out https://openalex.org I'm pretty optimistic that by the time google scholar really goes to shit they'll be good enough to pick up the slack.
60.
by smeeth 3y ago
Nope. Failure to maximize returns is not losing money.

More ›