nalzok

6 ms·

1.
by nalzok 2y ago
$ dig www.nyu.edu +noall +answer -t A ; <<>> DiG 9.10.6 <<>> www.nyu.edu +noall +answer -t A ;; global options: +cmd www.nyu.edu. 1 IN CNAME dsvdfvx64.github.io. dsvdfvx64.github
2.
Tutorials on Tinygrad(mesozoic-egg.github.io)
2 points by nalzok 2y ago|0 comments
3.
by nalzok 2y ago
I wonder what would happen if we quantized each dimension to 0.5 (or even fewer) bits instead of 1, i.e., taking 2 (or more) scalar components at a time and mapping them to 0 or 1 based on some carefully designed rules.
4.
by nalzok 2y ago
I agree that Nix is the right idea, but I'm hesitant to adopt it right now because (i) rough edges like poor documentation and inconsistent interfaces make it feel immature, (ii) things are moving very fast with experimental stuff li
5.
by nalzok 2y ago
You may want to check out [rcmd]( https://lowtechguys.com/rcmd/ ).
6.
by nalzok 2y ago
> by kqr, published 2024-11-19 It's from the future! ;)
7.
by nalzok 2y ago
I wonder if this has anything to do with Apple Intelligence. Maybe they can have an LLM that operates on encrypted input text without decrypting it, so that users can send sensitive information to an Apple-controlled central server without
8.
by nalzok 2y ago
> I began my research by looking for some kind of hard plastic case. While these are common for standard SD cards, they are quite rare for microSD. What about putting the microSD into an adapter [1] first? I imagine you can find a much b
9.
by nalzok 2y ago
How do you like Simutrans? https://www.simutrans.com
10.
by nalzok 3y ago
Congratulations on the release! How can we download the model and run inference locally?
11.
by nalzok 3y ago
You can install a free (as in beer) Wolfram engine on any computer, e.g. a beefy virtual machine: https://www.wolfram.com/engine/
12.
by nalzok 3y ago
> Subjects: Computation and Language (cs.CL); Information Retrieval (cs.IR) I'm surprised they didn't put `Machine Learning (cs.LG)` and `Machine Learning (stat.ML)`.
13.
by nalzok 3y ago
Is this equivalent to treating the linear recurrence as a convolution and computing it with a Fourier transform? See https://www.youtube.com/watch?v=OpJMn8T7Z34&t=840s
14.
by nalzok 3y ago
Nevermind, I figured it out: https://github.com/facebookresearch/metaseq/blob/main/projec...
15.
Model training diary/journal for LLMs?
1 points by nalzok 3y ago|1 comments
16.
by nalzok 3y ago
> How does that work? I don't really know, I pulled this out of the void with a SAT solver. Basically magic. That's pretty cool. How do you ask an SAT solver for "the nearest integer with the same population count", a
17.
by nalzok 3y ago
At the same time, the wallpaper says "Sipeed" instead of something mirrored like "beeqiS". Weird...
18.
by nalzok 3y ago
Can you share a link to your code repository? I want to play with your algorithm and observe the speedup on my own. Also I'm a big fan of JAX!
19.
by nalzok 3y ago
Wait, NUCs are that cheap? I would appreciate a link so that I can get my hands on one. I don't use eBay often and filtering feels terrible on their site.
20.
by nalzok 3y ago
> The main advantage is in the tinygrad IR. It has 12 operations, all of which are ADD/MUL only. `x[3]` is supported, `x[y]` is not. Can someone educate me why that is the case? Does `x[y]` require a Turing-complete kernel to comput
21.
by nalzok 3y ago
Seriously? I guess it's common sense among people living in the U.S, but as an international student I honestly didn't know it's legal to emulate a proprietary system.
22.
by nalzok 3y ago
Maybe make a pull request to the repo with what you said in the comment so that people like me don't have to ask this question in the first place? I assume this is a very frequently asked question.
23.
by nalzok 3y ago
I wonder why Nintendo hasn't attempted to take it down. It seems that the only use of such a "Nintendo Switch Emulator" is to play pirated games?
24.
by nalzok 3y ago
Could you explain what you mean by "Chinchilla under-trained" or "Chinchilla over-trained"? I assume it refers to some measure of trained-ness, but Googling yielded nothing relevant.
25.
by nalzok 4y ago
> at market price $600k I don’t think they are really spending $600k. Chances are that they own a GPU cluster so they don’t need to pay the cloud premium.
26.
by nalzok 4y ago
Do you have a plan to expose some high-level API endpoints? I have been dreaming about something like `run.colab.research.google.com/<notebook_url>?runtime=gpu` which executes a Colab notebook without human interference. This can
27.
by nalzok 4y ago
You should check out GreenArrays. Their G144A12 is an amazing little asynchronous chip. It's not an FPGA though.
28.
by nalzok 5y ago
What functions of Mathematica can do the job? I'm interested in checking it out.
29.
Ask HN: How do you debug yourself?
1 points by nalzok 7y ago|4 comments
30.
by nalzok 7y ago
I was waiting for him to share thoughts on the failure (?) of Lisp, which I believe is a decent "shopping mall", but unfortunately he did not :(

More ›