scrollop

14 ms·

1. 211.▲
   by scrollop 2y ago

   Also, is it a coincidence that at cheaper (potentially faster?) model has been released (just) before they roll out the "new" voice mode (which boasts very low latency)?
2. 212.▲
   by scrollop 2y ago

   From what I've learned from OpenAI, the "latest" "cheaper" model will perform worse than the previous model on various tasks (esp reasoning).
3. 213.▲
   by scrollop 2y ago

   Obesity rates have tripled in the us over the last 60 years- https://usafacts.org/articles/obesity-rate-nearly-triples-un... Please show me references substantiating your claim that obesity is "in partz due to hor
4. 214.▲
   by scrollop 2y ago

   As a primary care physician (and previous ED doctor) I don't think I'm qualified to go deep into the details of the vaccines - I trust my many more specialised colleagues around the world who unanimously support the vaccines and t
5. 215.▲
   by scrollop 2y ago

   As a primary care physician I have seen a huge number of long cover cases in the last 6 months. The people I see, of ages from 18 to elderly, are not having a continuous feeling of any sort of relief when I tell them their symptoms are like
6. 216.▲
   by scrollop 2y ago

   Which is perhaps why they have the anti stalking sound.
7. 217.▲
   by scrollop 2y ago

   Well, perhaps there will be a small revival, just as there has been with vinyl and film cameras, however compared to the "mainstream" these will likely be very small. If anything, I think it's more likely that people will inv
8. 218.▲
   by scrollop 2y ago

   I wonder if you could create an AI bot that interacts in the arena, and knows the telltale signs of a particular LLM, and when it sees that LLM answer, always votes for it. IF it does this hundreds of time, it could alter the results. Do AI
9. 219.▲
   by scrollop 2y ago

   +1
10. 220.▲
    by scrollop 2y ago

    IMO Lmsys benchamarks are essentially: How AI nerds (like us, so a particular subclass of the "normal" population) rate responses from various AIs - this rating is subjective, of course, and an LLM that produces answers that read
11. 221.▲
    by scrollop 3y ago

    Anthropic have released "tools" for Claude API- seems to be a way to essentially create plugins. https://youtu.be/6wkFb2_cUik?si=6gdpubgOiSUTlCAq (Not sure why there isn't a blog post or news on this yet, jus
12. 222.▲
    Claude API can now access the internet via "tools" that you can create(i.imgur.com)

    28 points by scrollop 3y ago|7 comments
13. 223.▲
    by scrollop 3y ago

    Here's a quick analysis of the model vs it's peers: https://www.youtube.com/watch?v=ReO2CWBpUYk
14. 224.▲
    by scrollop 3y ago

    So your theory would have it that if you repeated the question at temp 1 it would give the wrong answer more often than the correct answer?
15. 225.▲
    by scrollop 3y ago

    So did claude 3.0 with temperaure 0, whilst with temp 1 it got it wrong.
16. 226.▲
    by scrollop 3y ago

    Temperature 1 - It answered 1 sister: https://i.imgur.com/7gI1Vc9.png Temperature 0 - it answered 0 sisters: https://i.imgur.com/iPD8Wfp.png
17. 227.▲
    by scrollop 3y ago

    I have performed a few tests where I uploaded a few small pdfs, and chatGPT couldn't analyse the data. Perhaps it depends on the pdf.
18. 228.▲
    by scrollop 3y ago

    /s
19. 229.▲
    by scrollop 3y ago

    How do you reduce the severe oversharpening with iphones? Is there an app that can take footage without oversharpening?
20. 230.▲
    by scrollop 3y ago

    Doesn't seem to be able to reduce the iphones severe oversharpening. Would pay good money for an app that accomplish that.
21. 231.▲
    by scrollop 3y ago

    He hires smart people and listens to their ideas. He's still a douchebag.
22. 232.▲
    by scrollop 3y ago

    Here's the website that the stream is viewing: https://reddark.untone.uk/