matanyall

5 ms·

1.
by matanyall 8mo ago
It's so funny, I've never done a cost-benefit analysis of having "good monitoring" and then still not being able to figure out what broke and needing to pull in someone who doesn't need the monitoring at all because
2.
by matanyall 1y ago
Can we just have something that replaced the page with test datasets and Disney IP to poison the training data? Or maybe just embed it into the page itself, but hidden?
3.
by matanyall 2y ago
git add passions.xml; git commit -m "feat: commitments"; git push
4.
by matanyall 2y ago
Lol, I love it
5.
by matanyall 2y ago
+1 for fastAPI
6.
by matanyall 2y ago
Gambling has a similar addiction profile to cigarettes and other drugs, so why not have the same kind of labels on every bet and app, something like, "FanDuel is legally required to tell you that sports gambling has been shown to cause
7.
by matanyall 2y ago
Why do people do things? Doing things is falsely valued. (Am partially sentient collection of coral, YMMV)
8.
by matanyall 2y ago
Damn, y'all ok over there in Nixland?
9.
by matanyall 2y ago
$0.59 input, $0.79 output ( https://wow.groq.com/ )
10.
Llama3 on Groq(groq.com)
35 points by matanyall 2y ago|7 comments
11.
by matanyall 3y ago
This is a huge issue. What's the point of doing replication studies (which we need) if folks just don't care about the result. It's heavy handed, but I think every time a study fails to reproduce its results, every single pub
12.
by matanyall 3y ago
Groq Engineer here, I'm not seeing why being able to scale compute outside of a single card/node is somehow a problem. My preferred analogy is to a car factory: Yes, you could build a car with say only one or two drills, but a mod
13.
Ask HN: What are folks using for dealing with K8s manifests these days?
2 points by matanyall 3y ago|0 comments
14.
by matanyall 3y ago
Here we go again!
15.
by matanyall 3y ago
Yeehaw, here's to another year gone by!
16.
by matanyall 3y ago
Yeah, that's raw llama2 for you. Fine tuning, LoRa, that's how you get the nice responses.
17.
by matanyall 3y ago
Wow!
18.
by matanyall 3y ago
Easy! Write a compression algorithm that works as follows: if you detect the input is the first gb of Wikipedia, return nothing. Else return the compressed artifact of your compression algorithm of choice. To decompress, running the decompr
19.
by matanyall 3y ago
That's wild! Feels similar to some of the crypto scams going around wrt stable coins.
20.
by matanyall 3y ago
Are you saying that the opposite of the OP story is the ideal?