j_bum

13 ms·

1. 211.▲
   by j_bum 2y ago

   Lovely! Thanks for sharing. Would these patterns keep generating indefinitely?
2. 212.▲
   by j_bum 2y ago

   I’m not sure it makes sense to differentiate between energy spend while being “productive”, and energy spent whole trouble shooting and problem solving. After all, trouble shooting can be viewed as a productive thing. Interesting idea thoug
3. 213.▲
   by j_bum 2y ago

   Doesn’t sound out of context to me. It’s not just about economic battles. Did you read the final paragraph? > If war is what the U.S. wants, be it a tariff war, a trade war or any other type of war, we’re ready to fight till the end.
4. 214.▲
   by j_bum 2y ago

   http://archive.today/mU1yk
5. 215.▲
   by j_bum 2y ago

   Re: interpellation [0]. This was highlighted in a recent article about Mr. Beast [1]. Hadn’t considered this being a truth for her channel, but I can see what you mean. [0] https://en.m.wikipedia.org/wiki/Interpellation
6. 216.▲
   by j_bum 2y ago

   Wrong thread I think :)
7. 217.▲
   by j_bum 2y ago

   What a wonderful blend of fields for your PhD. I tend to agree with you, but I would be cautious in stating that an understanding of music is required for AGI. Plenty of humans are tone deaf and can’t understand music, yet are cognitively c
8. 218.▲
   by j_bum 2y ago

   Excellent point, and I agree with you. I’m curious if RAG can solve any of this problem?
9. 219.▲
   by j_bum 2y ago

   Agreed at every level. I absolutely love LLMs. I see them as insanely useful, interactive, quirky, yet lossy modern search engines. But they’re fundamentally flawed, and I don’t see how an “agent” in the traditional sense of the world can a
10. 220.▲
    by j_bum 2y ago

    Agreed, but I suppose this is a tell. I think they’re trying to place this into a separate class of models. I.e., we know it might not be as good as 3.7, but it is very friendly and maybe acts like it knows more things.
11. 221.▲
    by j_bum 2y ago

    Honest question for you: are these puzzles actually a good way to test the models? The answers are certainly in the training set, likely many times over. I’d be curious to see performance on Bracket City, which was featured here on HN yeste
12. 222.▲
    by j_bum 2y ago

    Any examples you’d be willing to share?
13. 223.▲
    by j_bum 2y ago

    What do you mean “heavier keys”? Each key (on a good piano) is weighted to have the same force requirements to be played. Of course varying forces can be used to achieve different decibels, but the curve is the same across the piano.
14. 224.▲
    by j_bum 2y ago

    Sure, but I’m curious if it would serve to provide some self-regulation. E.g., all of this “thinking” trend that’s happening. It would be interesting if the model does a first pass, scored its individual outputs, then reviews its scores a
15. 225.▲
    by j_bum 2y ago

    This is naive, but can you ask the model to provide a confidence rating for sections of the document?
16. 226.▲
    by j_bum 2y ago

    I completed my PhD in neuroscience at a mid-tier university. One piece of advice I can give you is to stay focused, but keep sight of the freedom and flexibility your time as a grad student will offer you. I was lucky to have an advisor tha
17. 227.▲
    by j_bum 2y ago

    Congrats, and good luck!
18. 228.▲
    by j_bum 2y ago

    An absolute word salad. Dead internet theory
19. 229.▲
    by j_bum 2y ago

    Great article. I share this sentiment and am glad I learned to code in grad school with only stack overflow as a spotty resource. We can definitely date OP by his use of “rawdog”, the context of which is actually hilarious and made me laugh
20. 230.▲
    by j_bum 2y ago

    I have aphantasia… so audio sports broadcasts are awful for me. Radio sports sound like a long string of facts that I need to remember but can get overwhelmed with. I’m jealous that radio is immersive for you!
21. 231.▲
    by j_bum 2y ago

    So expressively and beautifully written.
22. 232.▲
    by j_bum 2y ago

    Quora was the first website I blocked when I started using Kagi. Life is much better now without it dominating my search results.
23. 233.▲
    by j_bum 2y ago

    I love the idea of this tool, but there are serious issues with using LLMs to summarize articles and text. Re: Apple’s Notification Summary Debacle For example, this headline with a score > 5 is flatly incorrect. “China launches innovati
24. 234.▲
    by j_bum 2y ago

    +1 my LLM spidy senses were tingling. It’s the exclamation point in the first paragraph, the concise and consistent sentence structure, and the lack of colloquial tone. OP, no worries if you’re real. I often read my own messages or writing
25. 235.▲
    by j_bum 2y ago

    Yep. I think a default state of skepticism is an absolute necessity when working with these tools. I love LLMs. I agree with OP them expanding my hobby capacity as well. But I am constantly saying (in effect) “you sure…?” and tend to have a
26. 236.▲
    by j_bum 2y ago

    Safari as well
27. 237.▲
    by j_bum 2y ago

    Agreed. RIP Apollo. But I’m the better for it. HN is my only active social media account.
28. 238.▲
    by j_bum 2y ago

    Sorry to ruin your guilt free parade, but don’t forget that your charging is still producing emissions - just from a distance. Congrats on the EV though. I really want one, but it’s just not practical for the 400+ mile trips we regularly ne
29. 239.▲
    by j_bum 2y ago

    I love to hypermile my accord hybrid, but when my wife pulls the exact same mileage without doing anything special, it makes me feel so goofy. I’ll continue to hypermile for fun anyway lol
30. 240.▲
    by j_bum 2y ago

    Anyone looking to do 3D modeling in Python should absolutely check out PyVista.