yeahwhatever10

7 ms·

31.
by yeahwhatever10 2y ago
Are USB, HDMI, and Ethernet really the ports you want for your embedded system?
32.
by yeahwhatever10 2y ago
People like this?
33.
by yeahwhatever10 2y ago
The problem remains the size of the dataset. You aren't going to get large enough datasets in these specific domains.
34.
by yeahwhatever10 2y ago
Part of me thinks the meanness with which people interact with each other online makes people more hesitant to interact with others irl.
35.
by yeahwhatever10 2y ago
CUDA is one part, but another part of Nvidia's lead is their focus on bandwidth both memory and GPU-GPU communication. AMD dramatically falls behind Nvidia in training because of its terrible collective times (AllReduce, AllGather, etc
36.
by yeahwhatever10 2y ago
How exactly would they get sued for this?
37.
by yeahwhatever10 2y ago
The product demos show some features but obviously not everything. https://www.youtube.com/watch?v=rxKghrZU5w8
38.
How Uber Optimizes LLM Training(uber.com)
2 points by yeahwhatever10 2y ago|0 comments
39.
by yeahwhatever10 2y ago
I don't hate the product, but God I hate appeal to authority.
40.
by yeahwhatever10 2y ago
It's called survivorship bias.
41.
by yeahwhatever10 2y ago
Floorplanning algorithms and solvers already exist https://limsk.ece.gatech.edu/course/ece6133/slides/floorplan...
42.
by yeahwhatever10 2y ago
Why do they keep saying "superhuman"? Algorithms are used for these tasks, humans aren't laying out trillions of transistors by hand.
43.
by yeahwhatever10 2y ago
You think she stole a few?
44.
by yeahwhatever10 2y ago
Where is the performance data?
45.
by yeahwhatever10 2y ago
When will we get the “Mill” cpu?
46.
by yeahwhatever10 2y ago
How exactly are they going to measure the compute number?
47.
by yeahwhatever10 2y ago
It really just focuses on travel based questions, huh? The AI is definitely a millennial.
48.
by yeahwhatever10 2y ago
Down voting this comment is burying your head in the sand. Censorship has been and will continue to be one the primary political fights of the 20s. They just arrested the Telegram founder a few days ago...
49.
by yeahwhatever10 2y ago
Likely a lack of recently published code. Claude has been trained more recently.
50.
The Economy of Desire (2023)(madfeed.co)
3 points by yeahwhatever10 2y ago|0 comments
51.
by yeahwhatever10 2y ago
Damn the downvotes, this is a good joke!
52.
by yeahwhatever10 2y ago
You guys really got knock it off with adding “oogler” to everything.
53.
by yeahwhatever10 2y ago
The phrase you want to search is "constrained decoding".
54.
by yeahwhatever10 2y ago
The Americans is pretty banal and follows the basic script you would expect. It's good background noise if you want 80s nostalgia though.
55.
by yeahwhatever10 2y ago
When I was a TA I always did a second pass to make sure everything was even. It’s not that hard.
56.
by yeahwhatever10 3y ago
Can you explain how you got a 95% discount?
57.
by yeahwhatever10 3y ago
Did you watch the new Star Wars trilogy? How about the new Mulan? If you did and did not think they were hamfisted, then I think you have a much higher threshold than most.
58.
by yeahwhatever10 3y ago
That is how I felt when watching the documentary on Marc-Andre Leclerc, eventually the dice were going to roll against him and they did.
59.
by yeahwhatever10 3y ago
Yes that is why Frank agreed to betray Michael in the first place, same with Frank's right hand man Cicci.
60.
by yeahwhatever10 3y ago
Thanks for this write up. Your comment clears up a lot of the confusion I've had around these time series transformers. How does lagged features for an MLP compare to longer sequence lengths for attention in Transformers? Are you able

More ›