why_only_15

6 ms·

31.
by why_only_15 1y ago
taking into account all the impacts on society, uber is a substantial improvement on what came before. sometimes laws are bad and it is good when you break them
32.
by why_only_15 1y ago
doing a u32 compare instead of an f32 compare is not rust-specific or indeed CPU-specific.
33.
by why_only_15 1y ago
This trick is very useful on Nvidia GPUs for calculating mins and maxes in some cases, e.g. atomic mins (better u32 support than f32) or warp-wide mins with `redux.sync` (only supports u32, not f32).
34.
by why_only_15 2y ago
Probably more like richest 10%, which applies to most people in the US
35.
by why_only_15 2y ago
Random unrelated point: in a 100km radius circle between Atlanta and Augusta there are ~2,000,000 people (calculated using https://www.tomforth.co.uk/circlepopulations/ )
36.
by why_only_15 2y ago
I'd be pretty curious to get patio11's opinion on why #5 happened.
37.
by why_only_15 2y ago
my claim is not that it would take two days for such a pipeline to run but that it would take two days to make an NLP pipeline whereas an LLM pipeline would be faster to make.
38.
by why_only_15 2y ago
Assuming the 10M records is ~2000M input tokens + 200M output tokens, this would cost $300 to classify using llama-3.3-70b[1]. If using llama lets you do this in say one day instead of two days for a traditional NLP pipeline, it's wort
39.
by why_only_15 2y ago
how is this a plateau since gpt-4? this is significantly better
40.
by why_only_15 2y ago
True, though it's still enough nutrients to feed three people's entire caloric needs.
41.
Website Vibes(girl.surgery)
3 points by why_only_15 2y ago|0 comments
42.
by why_only_15 2y ago
What is this typically called? Are you referring to raw milk? I think this is a different thing.
43.
by why_only_15 2y ago
What he means is that 10M cows, all the cows in the US, produce dairy products that are 10% of calories in US diets.
44.
by why_only_15 2y ago
Fantastic article. I didn't realize dairy cows lactated ~60lbs/day or ~3.5% of their body weight. Totally insane. Chickens appear to be this way too -- from some quick research Rhode Island Reds are ~3kg, lay ~300 eggs/year a
45.
by why_only_15 2y ago
those cows probably still exist right? you can still get milk from them. Do they taste better?
46.
by why_only_15 2y ago
Used to really love his stuff, but this is all pretty well-trod ground and he doesn't have much new to say.
47.
by why_only_15 2y ago
I'm not too up to date but as I recall there are a lot of weirdnesses because of how big their chip is (e.g. thermal expansion being a problem). I believe they have a single giant line in the middle of the chip for this reason. maybe t
48.
by why_only_15 2y ago
they have about 125GB/s of off-chip bandwidth
49.
by why_only_15 2y ago
My understanding is that they mask off or otherwise disable a whole row+column of cores when one dies
50.
by why_only_15 2y ago
the cost is not the memory technology per se but primarily the wires. SRAM is fast because it's directly inside the chip and so the connections with the logic that does the work is cheap because it's close.
51.
by why_only_15 2y ago
Why would converting a specific LLM to an ASIC help you? LLMs are like 99% matrix multiplications by work and we already have things that amount to ASICs for matrix multiplications (e.g. TPU) that aren't cheaper than e.g. H100
52.
by why_only_15 2y ago
Not enormous without significant changes to the ML. There are two pieces to this: improving efficiency and improving flops. Improving flops is the most obvious way to improve speed, but I think we're pretty close to physical limits for
53.
by why_only_15 2y ago
The number for that is I believe 1 terabit or 125GB/s -- 21 petabytes is the speed from the SRAM (~registers) to the cores (~ALU) for the whole chip. It's not especially impressive for SRAM speeds. The impressive thing is that the
54.
by why_only_15 2y ago
Why does portability matter here? I guess if you want to run on macOS too?
55.
by why_only_15 2y ago
I'm confused about optimizing 7 watts as important -- rough numbers, 7 watts is 61 kWh/y. If you assume US-average prices of $0.16/kWh that's about $10/year. edit: looks like for the netherlands (where he lives) thi
56.
by why_only_15 2y ago
People typically quote google search as ~300k QPS
57.
by why_only_15 2y ago
To some degree this is true, but he’s also FOIA’d documents that describe what’s officially sanctioned.
58.
by why_only_15 2y ago
In the post he mentions why he thinks this is unlikely and is not a thing the US has done previously.
59.
by why_only_15 2y ago
They're not saying
60.
by why_only_15 2y ago
Specifically the clause is that you cannot use their consumer cards (e.g. RTX 4090) in datacenters.

More ›