https://arxiv.org/abs/2310.01889 (Submitted on 3 Oct 2023)
8,476 karma · joined September 7, 2018
https://arxiv.org/abs/2310.01889 (Submitted on 3 Oct 2023)
Anyways, it's the best JSON parser I found (in any language), I implemented fastgron (https://github.com/adamritter/fastgron) on top of it because of the on demand library performance.
One problem with the library was that it needed extra padding at the end of the JSON, so it didn't support streaming / memory mapping.
It's a great way for governments to hold BTC without raising suspicion of ,,advocating for'' it.
Nobody will be too surprised if the medicine will be retracted, as it was a very controversial approval process.
As it takes a lot of time to go through the process of changing things, make sure to invest time in deciding what that change should be (and how to structure it in a way that it's easiest to go through the process) instead of making the change that you would do if you would work on the code base yourself.
Just like Craig Wright forges documents and doesn't quite tell the truth, his description of the settlement offer isn't quite accurate either - it comes with loopholes that would allow him to sue people all over again.''
The crazy part is that Craig can still get out of all the trouble he caused if he can settle.... luckily Peter McCormack isn't settling.
An import tax on coal burning countries (especially China) would push the incentives in the right direction instead of outsourcing the climate externalities to them.
For machine learning speed compiling to the right CUDA / OpenCL kernel is much more crucial, so there's where the money goes.
Also you wrote that you suggested a change in the process, not that it was accepted.
I would definitely include Mamba state space models and of course would prefer a technical review over a corporate review of the year.
Svelte 5 builds upon these signals / derivations as well, and I guess a bunch of other libraries.
It would be time for them to get together and standardize on a signals library at this point that could be adopted to be a web standard and help interoperability between web components built with different frameworks.
Anybody who has worked with them knows that it's basically the same.
Optimizing kernels for all architectures and all models is a hard problem as there are lot of cases to handle, but getting good utilization for a particular model on a particular architecture (like MI300 / Mixtral) is not that hard.
Quite often it contains the experience of a life of a person condensed to a few hundred pages.
ChatGPT gives easier access to the knowledge contained in tens of thousands of these books. As for me I have been reading less and less books as more wisdom is accessable on the internet in better forms (now GPT).
I'm not against what OpenAI is doing as it moves humanity forward, but like you said I won't stop using ChatGPT just because ByteDance scrapes it.
I looked at the whitepaper and they didn't even mention it.
The reason e2e encryption got enabled is to make people feel safe.
I fixed the semicolon bug, but of course correctness is more important.
I'm thinking of doing some marketing (for example a blog entry just to show what was the main learnings in I/O and memory management in order to achieve this speed).
https://github.com/adamritter/fastgron/blob/main/src/print_g...