kadushka

15 ms·

91.
by kadushka 1y ago
The article is about staying in tech.
92.
by kadushka 1y ago
Why wouldn’t predicting the next frame in a video stream be as effective as predicting the next word?
93.
by kadushka 1y ago
Are you disputing the fact that cats are natural born predators?
94.
by kadushka 1y ago
Cats, like any other predators, are born to kill other animals. It’s a natural order of things. My cat regularly brings back mice, birds, lizards and little snakes. I do have to get rid of them, but he makes me proud.
95.
by kadushka 1y ago
Were you afraid you had it at 29?
96.
by kadushka 1y ago
Hiding things from parents is a far bigger problem than staring at a phone in a living room.
97.
by kadushka 1y ago
An output of a fab, just like an output of an LLM, is non-deterministic, but is good enough, or is being optimized to be good enough. Non-determinism is not the problem, it's the quality of the software that matters. You can repeatedly
98.
by kadushka 1y ago
Any laptop with an Nvidia card
99.
by kadushka 1y ago
What was different eight months ago?
100.
by kadushka 1y ago
Markus has been consistently wrong over the many years predicting the (lack of) progress of the current deep learning methods. Altman has been correct so far.
101.
by kadushka 1y ago
The funny thing is, if you asked “what is AGI” 5 years ago, most people would describe something like o3.
102.
by kadushka 1y ago
I just did it myself in the end.
103.
by kadushka 1y ago
Retraining the whole model from scratch every time you wanted it to learn something is not a solution. does a model need to retain its generality? Only if you want it to remain smart.
104.
by kadushka 1y ago
The most obvious blocker is catastrophic forgetting.
105.
by kadushka 1y ago
Yesterday I asked 2.5 Pro, Opus 4, and o3 to convert my Pytorch script from pipeline parallel to regular DDP (convert one form of multi-GPU execution to another). None of the three produced fully correct code. Even when I put together the 3
106.
by kadushka 1y ago
The ideas at Ogma are inspired by Numenta's work.
107.
by kadushka 1y ago
The reason I drank it is I like the taste. I am now using an RO filter, and the taste is not as good.
108.
by kadushka 1y ago
I drank only distilled water for 16 years. No supplements all that time, just regular food. No B12 deficiency or any other health issues.
109.
by kadushka 1y ago
inspired an entire genre of software-creating software. In this timeline, people shape their computing experiences as easily as one might sculpt a piece of clay, creating personal apps that make perfect sense to them and fit like a glove
110.
by kadushka 1y ago
They pivoted to regular deep learning when Jeff stepped away from the company several years ago. It does not appear they're doing much of brain modeling these days. Last publication was 3 years ago.
111.
by kadushka 1y ago
“Clearest images to date” - wouldn’t images taken from space, much closer to the sun, be clearer?
112.
by kadushka 1y ago
Wow, that ability is incredible! Thank you for sharing.
113.
by kadushka 1y ago
Coursera and Udacity have been around for a long time. Not to mention countless youtube lecture courses on any imaginable subject.
114.
by kadushka 1y ago
Did he break any laws?
115.
by kadushka 1y ago
I have done both physical labor and software engineering, so I’m speaking from personal experience. If your experience differs, what can I say - everyone is different.
116.
by kadushka 1y ago
Try physical labor: plumbing, electrical, construction. You might like it, and if not you’ll appreciate how easy you had it as a swe.
117.
by kadushka 1y ago
Only if computation is the bottleneck. GPT-4.5 shows it’s not.
118.
by kadushka 1y ago
He did go into specifics and explained his point. Or have you only read his first post?
119.
by kadushka 1y ago
what about o3?
120.
by kadushka 1y ago
Was your fix commented? If not, a human developer refactoring your code could have done the same. I’ve seen it happen more than once.

More ›