Live data from Hacker News

Viewing profile — crazyedgar

crazyedgar

HN member
Joined
Tue, Dec 20, 2022, 1:51 PM UTC
HN karma
41
Public activity
20 items

About crazyedgar

No profile information was provided.

Recent public activity

  1. comment
    Comment #35825790

    We just use https://www.sbert.net/ . Compare the embedding of the answer with the embeddings of YES versus NO.

  2. comment
    Comment #35825628

    There's no suspicion. Our experience shows that for the remaining 20% of work, LangChain actually gets in the way and must eventually be removed.

  3. comment
    Comment #35825507

    It's worse than that. The documentation is a confusing mess that completely omits the explanation of key default parameters and details. And the abstractions are horrendously britt…

  4. comment
    Comment #35825446

    I disagree. Production systems don't need to be full of AbstractSingletonProxyFactoryBeans which is basically what LangChain is. For example, Linux certainly isn't.

  5. comment
    Comment #35825064

    I like the idea, but I think a library that focuses on producing requests and parsing responses according to schema is better. Sending requests to the server is orthogonal to the p…

  6. comment
    Comment #35824989

    For us LangChain actually caused more problems than it solved. We had a system in production which after working fine a few weeks suddenly started experiencing frequent failures (m…

  7. comment
    Comment #35824935

    CozoDB? https://news.ycombinator.com/item?id=35641164

  8. comment
    Comment #35820689

    That's not what I meant by interaction. The evaluator had to ask the models to do tasks for them that they thought of by their own. Otherwise there are just too many ways that info…

  9. comment
    Comment #35820251

    This is my experience too. While I'd really love the Open Source models to catch up, currently they struggle even with dead-simple summarization tasks: they hallucinate too much, o…

  10. comment
    Comment #35820060

    This is hugely misleading. If your bot just memorizes Shakespeare and output segments from memorization, of course nobody can tell the difference. But as soon as you start interact…

  11. comment
    Comment #35819983

    Are you sure? I have yet to see any evidence that anyone at all (including Google) has built a model (or a "platform" as you prefer to refer to them) that can follow instructions a…

  12. comment
    Comment #35641319

    NVidia made this, OK. Not sure what is the significance of this? Can someone explain?

  13. comment
    Comment #35611669

    I believe this is the code for their work, which I believe the authors didn't put in the paper: https://github.com/philzook58/egglog

  14. comment
    Comment #35603280

    I think "creepy"

  15. comment
    Comment #35603202

    With "paywall" in the name, I don't think this is going to sell.

  16. comment
    Comment #35603149

    Why is there a waitlist on the front page? Is it not available yet?

  17. comment
    Comment #35603118

    No I don't think it is fine to ask for people's OpenAI key on a web page with obfuscated javascript. And apparently source code is not available.

  18. comment
    Comment #35603030

    Looks great!

  19. comment
    Comment #34066054

    Maybe it's intentional, to generate traffic.

  20. comment
    Comment #34065906

    I don't get the point of running it 24/7 as you can watch it on demand on youtube [1]. (And most of the people who are interested in it have already watched the whole thing already…