make3

25 ms·

301.
by make3 1y ago
maybe in 11 to 15 years after eventual respective Democratic and a non-Trump Republican governments
302.
by make3 1y ago
There were diplomatic treaties from Obama that worked, and that Trump cancelled in his last mandate. Doing things competently often allows one to get strictly better outcomes in life, and the current US regime is just horrible.
303.
by make3 1y ago
the OG of this were the google glasses, that got banned in restaurants etc. no such backlash this time around
304.
by make3 1y ago
it's not because everything sucks that we need to normalize making it worse
305.
by make3 1y ago
I don't think we should normalize pointing cameras at people's faces all the time. I hate these things.
306.
by make3 1y ago
not really, you can just run the blocks one by one like `nbconvert --to script` does. Looks like https://pypi.org/project/importnb/ exists
307.
by make3 1y ago
I feel like being able to import notebooks like `import notebook_name` and run jupyter notebooks (more easily) like `python notebook.ipynb` and the analogue in different languages would already get us 99% of the way there
308.
by make3 1y ago
it's ok if we both do it, not just one
309.
by make3 1y ago
not unless they're explicitly merged, which is not a requirement but a small speed only thing
310.
by make3 1y ago
agree to disagree
311.
by make3 1y ago
I feel like tu vs vous is extra annoying in French when you're speaking to someone who's not a stranger. I have no desire to judge if you think you're too famous of eg a professor for me to say tu, your success doesn't c
312.
by make3 1y ago
Using puzzles is not special or anything, it has been done a million times since before (and including) the LSTM paper (1997) https://www.bioinf.jku.at/publications/older/2604.pdf
313.
by make3 1y ago
in this case the LLM uses context clues and commonality priors to find the closest correct input, which is definitely relevant
314.
by make3 1y ago
this is not super constructive, is it really the top comment on HN
315.
by make3 1y ago
Are you saying that IronPython in general has memory leak issues? or just because you were trying to add support for an old .NET framework
316.
by make3 1y ago
I don't think people make the distinction like that. The open source vs non open source distinction boils down to, usually, can you use it for commercial use. what you're saying is just that it's non reproducible, which is a
317.
by make3 1y ago
this is pretty empty
318.
by make3 1y ago
really cool article.
319.
by make3 1y ago
plateau means stopped
320.
by make3 1y ago
the increases are not as fast, but they're still there. the models are already exceptionally strong, I'm not sure that basic questions can capture differences very well
321.
by make3 1y ago
it just makes it too easy to distill the reasoning into a separate model I guess. though I feel like o3 shows useful things about the reasoning while it's happening
322.
by make3 1y ago
like when Linda McMahon called AI A1
323.
by make3 1y ago
it's funny I feel like top of the line LLMs are basically AGI already or very close, you can have reasonable discussions with them about any subject etc & that a lot of anti LLM talk is grasping at straws & goalpost moving
324.
by make3 1y ago
maybe you can do something like speculative decoding where you decode with a smaller model until the large model disagrees too much at checkpoints, but use the context free cache in place of a smaller LLM from the original method. you could
325.
by make3 1y ago
It's not, this isn't a lawsuit, it's a casual discussion, and I don't really care whether I'm convincing you or not, I'm taking about this because I enjoy it. I can tell you though that DeepSeek R1 & RL app
326.
by make3 1y ago
do you see meaningful reasons for it to not get there or are you just making a snarky comment
327.
by make3 1y ago
that's not their best bet, their best bet is Gemini integration with all Google Workspace apps and Gemini eating Google search progressively
328.
by make3 1y ago
With the state of how the ai agents stuff already is, if you don't think that most tech jobs will be automated away in our lifetime, I don't what to tell you. Within 10-15 years, AI research automates itself and no one ever has a
329.
by make3 1y ago
I'm not sure hypocritical is the right word as you have no idea who parent is, maybe it's Pope Bob for all you know
330.
by make3 1y ago
people have been using language AI for advertising since Adsense

More ›