olmo23

6 ms·

61.
by olmo23 5mo ago
This is an active area of research. Demis Hassabis proposed training a model with a strict knowledge cutoff before 1915, and seeing whether it can independently arrive at general relativity.
62.
by olmo23 6mo ago
We are there. This is pretty much the reason why Mythos isn't being released publically.
63.
by olmo23 6mo ago
I don't think it would take very long to regain those skills either.
64.
by olmo23 6mo ago
Cropping doesn't necessarily remove exif. It doesn't even always remove the original pixels (simply setting cropTop and cropLeft and similar fields). My mind was blown when I saw rainbolt uncrop a picture. Anyways as mentioned els
65.
by olmo23 6mo ago
I think it uses exif data for that, because when I tried it did show my location but the vision api was overloaded, so nothing more showed up.
66.
by olmo23 6mo ago
A short reply without much substance
67.
by olmo23 6mo ago
I told my parents: if they are ever called by anyone, to tell them "now is not a good time, please give me a case number and I'll call back when I do have the time." And then, this is important, look up the number for the cus
68.
by olmo23 6mo ago
I agree the functions in a file should probably be reasonably-sized. It's also interesting to note that due to the way round-tripping tool-calls work, splitting code up into multiple files is counter-productive. You're better off
69.
by olmo23 7mo ago
I think he was referring to a cryptographic signature, possibly using the "web of trust" to get the key. I'm not convinced we need central authority to solve this.
70.
by olmo23 7mo ago
If you already know some php python javascript and/or c, you can pretty much just wing with Claude code.
71.
by olmo23 7mo ago
If you're already passing over the sources to strip the types, why would you also not do tree-shaking and minifications?
72.
by olmo23 7mo ago
It's so they can talk to their bot using Apple iMessage. That's pretty much it.
73.
by olmo23 7mo ago
Another great take I found online: "Don't send us a PR, send us the prompt you used to generate the PR."
74.
by olmo23 8mo ago
this is never going to be faster because it requires syscalls
75.
by olmo23 8mo ago
it would be prompted for the master password again, according to the website
76.
by olmo23 8mo ago
IIRC the official test-suite is not open-source, so I'm not sure how possible this is.
77.
by olmo23 9mo ago
Be realistic, demand the impossible.
78.
by olmo23 1y ago
You can not rely on brute force alone to compute these numbers. They are uncomputable.
79.
by olmo23 1y ago
In my experience most of the work a programmer does just isn't very difficult. LLMs are perfectly fine for that.
80.
by olmo23 1y ago
Racket is the canonical LISP (well, Scheme) to recommend to new users. https://racket-lang.org/
81.
by olmo23 1y ago
They gave us copilot and honestly for me it's been a bit of a double-edged sword. I often use it with Claude 4 as model. What I noticed is that the tool is able and eager to make large sweeping modifications to the code base where I wo
82.
by olmo23 1y ago
2.0 programs (model weights) are created by running 1.0 programs (training runs). I don't think it's currently possible to ask a model to generate the weights for a model.
83.
by olmo23 1y ago
Where does the knowledge come from? People can only post to SO if they've read the code or the documentation. I don't see why LLMs couldn't do that.
84.
by olmo23 1y ago
Just because humans can describe it, doesn't mean they can understand (predict) it. And the web contains a lot more than people's expressions: think of all the scientific papers with tables and tables of interesting measurements.
85.
by olmo23 3y ago
Racket
86.
by olmo23 3y ago
ChatGPT tells me "A kilogram of feathers and a kilogram of lead both weigh the same, as they each have a mass of one kilogram." (even though the question doesn't mention lead).