cavisne

7 ms·

61.
by cavisne 1y ago
The signal protocol is public, using their servers is frowned upon but its not a source code license violation.
62.
by cavisne 1y ago
From the paper it was a speedup on the XLA GPU kernel they wrote using Jax, which is probably not SOTA. I don't think Jax even has a official flash attention implementation.
63.
by cavisne 2y ago
"An H100 GPU has 989 TFLOPs of half-precision matrix multiply compute, and ~60 TFLOPs of “everything else”" I always thought there was a lot of crossover between gaming GPU's & DC GPU's (and the volume is why NVIDIA
64.
by cavisne 2y ago
MITRE has a trademark on the term CVE.
65.
by cavisne 2y ago
Hybrids are gas cars.
66.
by cavisne 2y ago
This is somewhat true but there is also a chasm where you earn good money but not enough to afford a mortgage. This chasm grows wider ever year in countries like Canada and Australia (and to a lesser degree America)
67.
by cavisne 2y ago
So now the finished products are tariff free, but the input costs when bought into America are tariffed. The exact opposite of what you would want to do to bring manufacturing back to America.
68.
by cavisne 2y ago
This is being sold as an air gapped product, it has to work offline by definition. Sure you could hide some way of phoning home and deploy it into the SCIF, but would you really want to risk a firing squad to improve some advertising metric
69.
by cavisne 2y ago
Seems pretty high, this is an air gapped product so at some point the employees of whatever government they are giving it to would need to SSH into the VM's to load new weights etc. Lots of ways to make it tricky/watermark the wei
70.
by cavisne 2y ago
The raw computation is just a bunch of matrix multiplications in a row, most of the algorithmic complexity/secret stuff would be around scaling & efficiency. For training the model the HW is much more important as you need to scale
71.
by cavisne 2y ago
Europe is mostly "old money" so this makes sense. They didn't earn that money they are custodians of it.
72.
by cavisne 2y ago
Factually despite all the noise there are no significant tariffs in either direction at this point in time, 98% of trade is tariff free under the free trade agreement. Mexico and to a lesser degree Canada are in a more favorable position th
73.
by cavisne 2y ago
In addition they use their government work to increase private sector profits, either implicitly "we work with your government" or explicitly (PWC offering briefings on tax policy they were helping with to the targets of the tax p
74.
by cavisne 2y ago
None of that onboard stuff ended up working afaik as the Tensor chips shipped right as giant LLM's became how dominant.
75.
by cavisne 2y ago
"In addition to the use of Signal, some DOGE staffers are bypassing other vetting processes and chains of custody for official government documents by working simultaneously out of Google Docs instead of circulating single copies of dr
76.
by cavisne 2y ago
You can do the same from the browser on any phone (including iphone) by going through AI Studio https://aistudio.google.com/live
77.
by cavisne 2y ago
Was that gamble wrong? I thought all LLM training workloads do collectives that involve all nodes (all-gather, reduce-scatter).
78.
by cavisne 2y ago
These complex 4D chess strategies never really pan out. Maybe someone in the administration thinks that was the goal but not Trump. During the campaign Trump said he wanted 10 percent tariffs on the world and 60 percent tariffs on China. Af
79.
by cavisne 2y ago
The JAX docs have a good explanation for how a TPU works https://docs.jax.dev/en/latest/pallas/tpu/details.html#what-... Its not really useful for other workloads (unless your workload looks like a bunch
80.
by cavisne 2y ago
Nvidia has ~60% margins in their datacenter chips. So TPU's have quite a bit of headroom to save google money without being as good as Nvidia GPU's. No one else has access to anything similar, Amazon is just starting to scale thei
81.
by cavisne 2y ago
"that kind of thing is going to come to America... and be automated". Pretty disingenuous cut on the quote there by the bluesky post.
82.
by cavisne 2y ago
The H100 theoretical flops number is just marketing, as it relies on sparsity that LLMs don’t use
83.
by cavisne 2y ago
Billboards being rare outside of the US seems quite incorrect. The developing world is full of billboards, and places in Europe like Milan have some wild Samsung billboards.
84.
by cavisne 2y ago
Yes this was the case for Venezuela, they started taking flights again when the El Salvador news came out. Refusing to repatriate citizens was part of the argument the Trump admin used to justify use of the Alien Enemies act. If this holds
85.
by cavisne 2y ago
Nearly 25% of El Salvador is in America, although they fled gang violence not Bukele.
86.
by cavisne 2y ago
In the presentation they said eventually kernels can share SIMT (PTX) and TileIR but not at launch. It seems pretty mysterious why they don't just emit PTX, I would guess they are either taking the opportunity to clean things up for ML
87.
by cavisne 2y ago
Scaling transformers has been basically alchemy, the breakthroughs aren’t from rigorous science they are from trying stuff and hoping you don’t waste millions of dollars in compute. Maybe the company that just tells an AI to generate 100s o
88.
by cavisne 2y ago
This article was prescient enough that I had to check in wayback machine. Very cool.
89.
by cavisne 2y ago
The problem is the hardware not the software, and specifically not CUDA. Triton for example writes PTX directly (a level below CUDA). Trying to copy Nvidia hardware exactly means you will always be a generation behind, so they are forced to
90.
by cavisne 2y ago
Glenn greenwald comes to mind.

More ›