anxman

5 ms·

1. 31.▲
   by anxman 2y ago

   Cryptic errors that will make you tear your hair out and docs that are constantly changing
2. 32.▲
   by anxman 2y ago

   ComfyUI
3. 33.▲
   by anxman 2y ago

   I'm not surprised that llama 3.2 11b failed here. The author should really have tested Llama 3.2 90b, which does remarkably better: Same query and image: This line graph illustrates the working-poor rate of the US labour force from 19
4. 34.▲
   by anxman 2y ago

   I'm really happy with Takeout. It has its flaws but it's the easiest stack to get up for web/ios/android.
5. 35.▲
   by anxman 2y ago

   Can we get a trial to test this out? Been looking for this.
6. 36.▲
   by anxman 2y ago

   TIL there’s a downvote on here
7. 37.▲
   by anxman 2y ago

   Those were still quantized right? My understanding is that LLaMA3 does not quantize well at all.
8. 38.▲
   by anxman 2y ago

   I also hate the billing. It feels like configuring AWS more than calling APIs.
9. 39.▲
   by anxman 2y ago

   And the training samples are overly tied to Vertex
10. 40.▲
    by anxman 2y ago

    Founder Preferred is a special class of stock that can convert into Preferred when sold. It’s different from Common as it doesn’t affect the 409A. IANAL.
11. 41.▲
    by anxman 2y ago

    I used Founders Preferred shares to get liquidity at the A (for a now defunct startup). In our case, we offered all vested employees the option of selling in the same round on the same terms. I personally don’t recall any disclosure require
12. 42.▲
    by anxman 3y ago

    This is my experience too. You have to be have deep domain knowledge to really get the LLM to do what you want. Then it saves me a ton of time.
13. 43.▲
    by anxman 3y ago

    Anyone know the karma threshold or how they're determining most active users?
14. 44.▲
    by anxman 3y ago

    This could be a product. AI sales training.
15. 45.▲
    by anxman 3y ago

    I bet WOTC would find this very useful!
16. 46.▲
    by anxman 3y ago

    I have a 15 year Prime account. I purchased fake Colgate toothpaste on Amazon and left a review to alert others. Amazon then deleted all reviews I’ve ever left and permanently banned me from leaving any new ones. I appealed twice with no re
17. 47.▲
    by anxman 4y ago

    I can confirm knowledge of this as well
18. 48.▲
    by anxman 4y ago

    This will end with a huge settlement. Ebay did wrong by targeting their most loyal merchants with a criminal campaign.
19. 49.▲
    by anxman 4y ago

    I received fake toothpaste from Amazon. The toothpaste gave me asthma and the packaging and consistency were different from store bought. I left a review stating the item I received was fake and then the review was removed by Amazon. Amazon
20. 50.▲
    by anxman 4y ago

    I worked on Sundars team at Google in 2006. He is highly technical and understands the nuance of software.
21. 51.▲
    by anxman 4y ago

    Holy moly
22. 52.▲
    by anxman 5y ago

    The real reason was that YouTube had become the number two search engine on the Internet. Nobody in Corp dev acquires a company simply because of who invested in it.
23. 53.▲
    by anxman 5y ago

    No need to apologize, it's precisely this double speak that Preston (and Peskin) use to advance their own political agendas at the expense of vulnerable communities here in SF. There is also a component of semi legally siphoning off p
24. 54.▲
    by anxman 5y ago

    It doesn't remove the images though
25. 55.▲
    by anxman 6y ago

    This is more like the NYPostTimes
26. 56.▲
    by anxman 6y ago

    This is called Google One. I pay for it.
27. 57.▲
    by anxman 6y ago

    Same here and I worked on Stadia
28. 58.▲
    by anxman 6y ago

    Nice article from Kim Mai
29. 59.▲
    by anxman 6y ago

    What is this?
30. 60.▲
    by anxman 6y ago

    Congrats. Great product.