glouwbug

6 ms·

91.
by glouwbug 4mo ago
I always wanted to wrangle ~10 of these together and practice distributed computing for the price of ~$200. That was a 2014 dream. I guess instead someone else ran with all of Jensen's polygon drawers and Hynix's backlog. Que sera
92.
by glouwbug 4mo ago
You can triangulate. Ask it the same thing in different ways and with different LLMs. Operate in domains where the output is verifiable, like in the sciences but in terms of numerical computing. Study the output, graph it, learn it, reason
93.
by glouwbug 4mo ago
You're literally able to write this because of all the obscure shit
94.
by glouwbug 4mo ago
Looks like the RPI5 is like $200 in Canada now. We might have lost the plot
95.
by glouwbug 4mo ago
The GOAT. I remember playing with CUDA in there around 2011/2012. GNU Octave if you’re on NIX or you don’t have $40k for a license
96.
by glouwbug 4mo ago
The whole point is that the layperson is not classically trained to know right from wrong which is the entire thesis of knowledge share. Even post doctorate students are required to have their work peer reviewed. It’s why anthropic and Open
97.
by glouwbug 4mo ago
COVID will be seven years old this December. Many of us here are still working from home since that time. It doesn’t feel like seven years. 2020 feels like last year. What can one typically accomplish in seven years? An undergrad, masters,
98.
by glouwbug 4mo ago
Those low quality lawyers and doctors are still vastly more capable than a layperson at verifying AI output
99.
by glouwbug 4mo ago
https://www.digitalmars.com/articles/C-biggest-mistake.html
100.
by glouwbug 4mo ago
C's biggest mistake. But in other news most don't know that a[3] == 3[a]
101.
by glouwbug 5mo ago
To be fair, it hasn’t really been one since SEO
102.
by glouwbug 5mo ago
People spend that much a month on restaurants
103.
by glouwbug 5mo ago
This is a good example of what many private companies are doing, and the rude awakening they’re in for with token price hikes and vendor lock in. It’s like Oracle all over again
104.
by glouwbug 5mo ago
Sometimes programs work because they rely on undefined behaviour. The users validate that. A famous quote: “we don’t break user space”
105.
by glouwbug 5mo ago
It’s just linear algebra. Work your way from feed forward to CNN to RNN to LSTM to attention then maybe a small inference engine. Kaparthy’s llama2.c is only ~300 lines on the latter and it pragma simds so you don’t need fancy GPUs
106.
by glouwbug 5mo ago
Probably because (*p).field already worked
107.
by glouwbug 5mo ago
Relatively speaking, a company could obtain AGI first merely by just by keeping their critical thinkers while the rest of the industry offloads their thinking to LLMs. It wouldn’t be AGI, just GI, like we’ve always had before before GPT arr
108.
by glouwbug 5mo ago
It’s about what your average senior engineer makes here at hackernews per month
109.
Show HN: Nibble(github.com)
102 points by glouwbug 5mo ago|25 comments
110.
by glouwbug 5mo ago
Classic no Scotsman
111.
by glouwbug 5mo ago
Sometimes entire businesses collapse to a python dict and a backup UPS power supply
112.
by glouwbug 5mo ago
Right, but it takes one to know one. Many don’t have the ability to decipher what’s good stable output or not
113.
by glouwbug 5mo ago
I’m going to say there’s some mixup of causation and correlation here
114.
by glouwbug 6mo ago
Funny how AI took all the jobs, but not from automation
115.
by glouwbug 6mo ago
That’s okay. I’m there with you too with about the same cumulative. I measure my own projects by the enjoyment I got out of them. No sense in chasing validation from others when ones only metric will forever be what’s in their own control.
116.
by glouwbug 6mo ago
Hackernews sees skewed burger and proceeds to praise itself
117.
by glouwbug 6mo ago
I had exactly this way of thinking last year and began specializing myself: github.com/glouw/ensim4 I reckon moving forward software will became an applied tool to the applied sciences. I mean, it always has been, but the barrier
118.
by glouwbug 6mo ago
The majority of humans are average, as is the training set
119.
by glouwbug 6mo ago
Ceiling and floor casting is tricky, and pretty expensive at times. Getting vertical panning to work too is tricky but well worth it.
120.
by glouwbug 6mo ago
With a couple of function pointers you can climb precedence with just functions: parse_left_to_right(with(), is_token()) { left = with() while(is_token()) { right = with() left = operate(left, right, operator) }

More ›