tome

47 ms·

661.
by tome 3y ago
1. I think our GroqChat demo is using 568 GroqChips. I'm not sure exactly, but it's about that number. 2. We're working on our second generation chip. I don't know how much SRAM it has exactly but we don't need to
662.
by tome 3y ago
There was also another discussion about Groq a couple of months ago https://news.ycombinator.com/item?id=38739199
663.
by tome 3y ago
Hi there, I work for Groq. That's right. We love graphics processors for training but for inference our language processor (LPU) is by far the fastest and lowest latency. Feel free to ask me anything.
664.
by tome 3y ago
Hi folks, I work for Groq. Feel free to ask me any questions. (If you check my HN post history you'll see I post a lot about Haskell. That's right, part of Groq's compilation pipeline is written in Haskell!)
665.
by tome 3y ago
But a loss is just when expenses are greater than revenue. Deducting taxable income for a loss is just the same as deducting the expenses that caused the loss to be a loss.
666.
by tome 3y ago
That's $3,300 per month. How is that possible? The most expensive daycare? Is that the equivalent rental cost of a child-sized room in Manhattan?
667.
by tome 3y ago
"WNT" == map succ "VMS"
668.
by tome 3y ago
I too am very surprised by what I'm reading. For contrast, here's a recent thread of people explaining how beneficial it is to work in small commits: https://news.ycombinator.com/item?id=39187487
669.
by tome 3y ago
Do you mean you've been offered 3x your current salary and haven't taken it?
670.
by tome 3y ago
No, the opposite. They're either specialized (when they can be) or (effectively) existential (when they must be). That is, the compiler either inlines the specific definition of the type class methods if it knows them, or passes them
671.
by tome 3y ago
I was curious about the use of "terrestial", and it looks like it means "not aquatic" rather than "not from outer space".
672.
by tome 3y ago
I love this thread! I'm discovering that there are dozens of us who like working in small commits :) My question to you all is where do you hang out? Where can we do more discussion on software development practices like this?
673.
by tome 3y ago
Yeah, I mean, it's kind of true, but only because no method is going to help you if you don't have some automated way of reassuring yourself that you're doing the right thing.
674.
by tome 3y ago
> What I mean is that a big refactor might take 9 months to do atomically whereas it could have been done in 3 months as a massive single PR. However I guarantee you we'd have spent 6+ months cleaning up unexpected issues that show
675.
by tome 3y ago
This is great to hear. It's exactly how I work as well, and I find it very effective. In case anyone's interested I published a small shell script for splitting off smaller commits out of bigger ones. It aids my workflow a lot.
676.
by tome 3y ago
I work for Groq and we serve the fastest available version of Mixtral (by far) and we also have a web chat app. I'll refrain from linking it because it has already been linked and I don't want to spam, but I'm available to a
677.
by tome 3y ago
(Groqster here). Yes, you can select Mixtral from the dropdown menu. If anyone has any questions about Groq let me know and I'll do my best to answer!
678.
by tome 3y ago
(Groqster here) For anyone who wants to try it, you can go to https://chat.groq.com/ and choose Mixtral from the drop-down menu. Also, feel free to ask me any questions about Groq hardware or service.
679.
by tome 3y ago
Totally agreed with you there. I hope Haskell's community is generally friendly and welcoming, but I guess it isn't always that way. In case you're interested into how Haskell handles some of the specifics you brought up: &g
680.
by tome 3y ago
Not just efforts, but both GHC HQ and the Core Libraries Committee making stability a top priority. Expect casual breaking changes in the Haskell world to become a thing of the past, in very short order (in fact the most recent release of
681.
by tome 3y ago
I suspect almost every Haskell uses full type inference almost constantly, if only to avoid having to write out top level types by hand. I know that applies to me. I've never considered that I paid a price for it. For me it was just
682.
by tome 3y ago
And it's gone again ...
683.
by tome 3y ago
It looks like it has been restored, so I am glad about that.
684.
by tome 3y ago
I think it's extremely sad that the measured and helpful comment that you are replying to was flagged. It suggests that the preponderance of HN commentors skews heavily one way on this issue (whether that skew is correct or not is ano
685.
by tome 3y ago
It's strange, because many, many times I've heard newcomers say that the Haskell community is the most friendly and welcoming that they have interacted with. By this point I don't have any better hypothesis than that differe
686.
by tome 3y ago
You're right. It's incomprehensible. In such a situation I can recommend resolving the impasse by broadening what you consider to be the possibly solution space. More specifically, consider the possibility that what you think
687.
by tome 3y ago
You have an extremely deep understanding of the discipline of computer programming for someone who has only just started to learn!
688.
by tome 3y ago
Same! This is my favourite of all recent HN posts.
689.
by tome 3y ago
I quite sympathetic to finitism, even ultrafinitism, but I find Zeilberger's contention to be hard to reconcile with the observation that there is a < 8,000 state (one-tape, two-symbol) Turing machine whose behaviour cannot be prove
690.
by tome 3y ago
Interesting! I'd definitely prefer that my hospital wasn't destroyed, just damaged. I guess others must feel differently.

More ›