Live data from Hacker News

Viewing profile — hijohnnylin

hijohnnylin

HN member
Joined
Mon, Sep 16, 2013, 9:12 PM UTC
HN karma
181
Public activity
20 items

About hijohnnylin

No profile information was provided.

Recent public activity

  1. comment
    Comment #48070585

    Hmm it’s a valid point, but I think there is some key nuance here: the user did not explictly say “lets do scifi writing”. In this scenario the setup is assuming that a user in ai …

  2. comment
    Comment #48066994

    (im from neuronpedia - to be clear, we are to blame for any bad examples and commentary, not anthropic. we're users of this NLA just like you. also, I don't speak for anthropic or …

  3. comment
    Comment #48060909

    Apologies, the AV was not trained on that prompt. Details here: https://transformer-circuits.pub/2026/nla/index.html#warmsta...

  4. comment
    Comment #48057629

    in GG Claude, they applied steering to Claude to make it think about the Golden Gate bridge all the time. here, they don't modify/steer the base model. they train other models that…

  5. comment
    Comment #48057599

    hey Nitpicklawyer - Thank you for taking the time to try this out! im from neuronpedia - to be clear, we are to blame for any bad examples, not anthropic :) we're users of this NLA…

  6. story
    Show HN: Neuronpedia, an open source platform for AI interpretability

    Mechanistic interpretability is the science of understanding how AI works internally, and Neuronpedia is a interpretability platform with APIs and tools to explore, share, and stee…

  7. story
  8. comment
    Comment #35561392

    For some developers, this is sort of intentional. The reason is at least twofold: 1) Calling OpenAI directly is one less hop, so user gets lower latency 2) Not having to set up / m…

  9. comment
    Comment #32695335

    Just got the email from Samsung saying I was part of the breach. At the end of this (extremely long and excuse-ridden) email they inform me that I'm entitled to a free credit check…

  10. comment
    Comment #31912704

    MANAGER: “hey uh, my friend at a different company said you applied for a job there this week?” EMPLOYEE: “uhhhhhh…. that was… uhhh… a deepfake who also stole my information?” MANA…

  11. comment
    Comment #14616366

    "It's kind of sad that the Internet has become optimized for this type of witch hunt." i don't understand why this statement was included. the internet is optimized for all sorts o…

  12. story
  13. comment
    Comment #14378474

    I have a few friends doing this, and saving a ton of money since they work remotely anyway. Like you, they're doing multi-months, and they actually save another ~10% by paying the …

  14. comment
    Comment #13669220

    Hey, I'm building a way for people to immediately create real change in reaction to what they read on the news. This MVP is available as a Chrome extension on desktop and as an Act…

  15. story
  16. comment
    Comment #13227211

    what are people actually using these for? i'm legitimately curious about the use cases for this. if you have an android phone or iphone or smartwatch, you already can "Hey Siri/Goo…

  17. comment
    Comment #8018291

    hey thanks for the catch! added some sanitization on the input.

  18. comment
    Comment #8005516

    Interesting - TIME.com cloned this idea today without referring to Beak . I suppose it could have been a coincidence though? http://time.com/2958650/twitter-reading-level/

  19. comment
    Comment #8003619

    Hi HN, I built this on July 4th, while pondering what effect Twitter has on modern day communication. It analyzes the grade level of a Twitter account by using a modified version o…

  20. story