Thread

1. olwmc · 1790653296 · focus · HN ↗

   Sorry, do we have actual clear implementation details for Jev? I keep seeing these "recreations" or "Do Jev at home" but do we have access to their architecture? I haven't even used the product, I just find it strange.

   1. Zetaphor · 1790654055 · focus · HN ↗

      What they did was immediately obvious and trivial to replicate

      1. prodigycorp · 1790657624 · focus · HN ↗

         They make it clear that their edge is in their synthetic data. Train your own jevs miss the point. And if you have that much data, you didnt need jev or these replacements in the first place.

         HN's desire to pretend the data pipeline doesn't exist or isnt meaningful is silly.

         1. Zetaphor · 1790657980 · focus · HN ↗

            I'd argue that a tiny fine-tuned Jev on my dataset that I can run for practically free is far more valuable if the idea is to make this a dependency in real work

            1. prodigycorp · 1790658539 · focus · HN ↗

               People have been making their own calibrated classifiers and decision makers for a very long time. Mine outperform jev. But that's not the point of jev.

               1. dcl · 1790659471 · focus · HN ↗

                  What is...?

                  1. ad1ttya · 1790661741 · focus · HN ↗

                     i guess it's not needing to finetune for each specific use case

                     1. jmpeax · 1790664212 · focus · HN ↗

                        Then the probabilities are not calibrated. Jev is basically the LLM version of the "is-even" library.

Unofficial Hacker News client; not affiliated with Y Combinator.