Live data from Hacker News

Viewing profile — esafak

esafak

HN member
Joined
Fri, Apr 09, 2021, 7:50 AM UTC
HN karma
8,365
Public activity
6,671 items

About esafak

Machine learning engineer by trade.

Recent public activity

  1. comment
    Comment #49250094

    That reductive analogy does not begin to describe the lengths GPT went to. Its task was to access a database file that had accidentally not been placed inside the model's container…

  2. comment
    Comment #49249253

    No, they are not. https://www.usinflationcalculator.com/inflation/airfare-infl...

  3. comment
    Comment #49248974

    This is not an issue. IF you prompt it right you could have avoided this with models at least a year old.

  4. comment
    Comment #49248832

    What is different about yours?

  5. comment
    Comment #49248696

    You can tell one to rewrite complex applications in a different language, solve open research problems, or develop novel viruses. This is nothing like pressing the gas pedal.

  6. comment
    Comment #49246240

    You imagine robots are going to play second fiddle to models, but they could become integral parts of model training with all the data they collect. Training could even become dist…

  7. comment
  8. comment
    Comment #49243981

    Free models help move robots; a complement. I can foresee the models staying free.

  9. comment
    Comment #49243779

    This is a great demonstration in the value of approximation. The IKEA is close enough for most people's tastes at a much more attractive price point.

  10. comment
    Comment #49238247

    Agreed. I created a ticket. (Not my product.)

  11. comment
    Comment #49237357

    LLMs are already slow on one machine. Now you are going to add network calls to it?

  12. comment
    Comment #49237085

    It might be a joke in reference to the video's title, "I loved every minute of it, however hard it had been".

  13. comment
    Comment #49237030

    https://en.wikipedia.org/wiki/Effective_accelerationism

  14. comment
    Comment #49234375

    The argument's so moronic I can't even follow it...

  15. comment
    Comment #49234117

    Open weights deter capex? Who cares!? Nvidia isn't paying me to increase Openai's capex. I want tokens, not capex. Am I misconstruing the argument?

  16. comment
  17. comment
    Comment #49232094

    Keeping Master Green at Scale https://dl.acm.org/doi/10.1145/3302424.3303970

  18. comment
    Comment #49224122

    > When AIs learn from other AIs, model collapse occurs. I don't think is so. Models continuing to improve by learning from other models is a counterexample.

  19. comment
    Comment #49221768

    I think we are in need of Europe's leadership in safety legislation. It is foolish to say 'China will get ahead' when they will harm themselves too. Being unsafe is not something t…

  20. comment
    Comment #49221725

    They also like weapons.

  21. comment
    Comment #49218030

    Reminds me of the time Cheney shot his friend while quail hunting, and his friend said he was "deeply sorry for all that Vice President Cheney and his family have had to go through…

  22. comment
  23. comment
    Comment #49216016

    tally.so also got affected. https://www.metabase.com/blog/security-update

  24. comment
    Comment #49215943

    Tell us more.

  25. comment
    Comment #49215569

    https://artificialanalysis.ai/?cost=intelligence-vs-cost-per...