Live data from Hacker News

Viewing profile — z991

z991

HN member
Joined
Mon, Apr 16, 2018, 3:53 PM UTC
HN karma
549
Public activity
39 items

About z991

No profile information was provided.

Recent public activity

  1. story
  2. comment
    Comment #44361604

    Their YouTube channel is equal parts fascinating, terrifying, and boring: https://youtube.com/@uscsb

  3. story
  4. comment
    Comment #43319712

    There's an important detail here -- an advantage of mRNA vaccines is that mRNA does not last all that long in cells [1]. Thus, even if the cells live a long time, they won't produc…

  5. comment
    Comment #41837959

    Wow that is also very cool. For those wondering, this is what it looks like: $ sudo apt install slop $ slop 1719x1403+1080+277

  6. comment
    Comment #41837644

    Built it and took a fullscreen screenshot with GIMP to figure out the width/height/x/y coordinates I wanted and tested with Google Meet. Working perfectly!

  7. comment
    Comment #41837564

    Wow, this is fantastic! This exact use case, on Linux, is why our company selected Zoom instead of Meet. Awesome!

  8. comment
    Comment #41099544

    Wayne Hale is always my favorite source for these: https://waynehale.wordpress.com/2014/10/26/sts-93-we-dont-ne...

  9. comment
    Comment #38834168

    I commend the authors on making this easy to try! However it doesn't work very well for me for general voice cloning. I read the first paragraph of the wikipedia page on books and …

  10. comment
    Comment #38440619

    You have a custom prompt enabled (probably from viewing another one and pressing "start over") that is asking for opposites which will increase the noise a lot.

  11. comment
    Comment #38437636

    The entire thing is frontend only (except for the share feature) so the server never sees your key. You can validate that by watching the network tab in developer console. You can …

  12. comment
    Comment #38437568

    No, in that case there is a custom prompt (visible in the top dropdown) telling GPT4 to replace everything with corgis when it writes a new prompt.

  13. comment
    Comment #38437332

    https://dalle.party/?party=RqpIijhH

  14. comment
    Comment #38436628

    Wow the empty prompt does much better than I'd have guessed

  15. comment
    Comment #38436234

    You can try a custom prompt and see if you can get GPT4V to stop doing that / if it matters.

  16. comment
    Comment #38436219

    Yeah, that's a bug, I'll try to fix it tonight!

  17. comment
    Comment #38433605

    Also, descent into Corgi insanity: https://dalle.party/?party=oxXJE9J4

  18. story
    Show HN: A Dalle-3 and GPT4-Vision feedback loop

    I used to enjoy Translation Party, and over the weekend I realized that we can build the same feedback loop with DALLE-3 and GPT4-Vision. Start with a text prompt, let DALLE-3 gene…

  19. comment
    Comment #38024951

    Summary: https://pdf2gpt.com/?summary=a60cdd709ca243599185af0f8513e33...

  20. comment
    Comment #37915981

    Actual paper: https://arxiv.org/abs/2310.10631 Summary: https://pdf2gpt.com/?summary=ecb8a970ea5c4a41ab985eca15b7740...

  21. comment
    Comment #37778066

    Summary: https://pdf2gpt.com/?summary=3ade1d183ee74c0aa8fff901e892dc5...

  22. comment
    Comment #37714941

    The actual research article: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4573321 Summary: https://pdf2gpt.com/?summary=84ff84d4b98b4f0c985a17d07db482c...

  23. comment
  24. comment
    Comment #36455176

    You might like pdfpc: https://pdfpc.github.io/

  25. story
    Show HN: Realtime LLM Chat on an 8GB Nvidia GPU

    Demo runs on a laptop 3070 Ti / 8GB. GPU memory doesn't go above 6GB, so it might run on an even smaller GPU. Uses a 4-bit 7bn parameter alpaca_lora model and performance is signif…