Lol, I love that framing. Yeah, the small models have impressed me a lot during this work. The reasoning can be quite good, and definitely sufficient for a lot of cases. Just gotta nudge em back on track Every now and then and they'll figure it out.
replyIf I understood correctly, the model will get it right because it knows when it isn't right.
replyEssentially, yes that's right! There's some subtlety in how to let it know it was wrong (returning things as tool errors because it trained on that), but that's the gist of it - sort of a self-correcting architecture.
replyreply
I was expecting this https://knowyourmeme.com/memes/the-missile-knows-where-it-is
replydeleted
replyA thousand monkeys on a thousand typewriters…
replyThat is the whole challenge, actually! A new metric I'm going to dogfood into forge is ETTWS - estimated time to working solution.
replyA simple retry loop around your whole workflow could, in some cases, be all you need. But it could mean many blind attempts to get through a workflow successfully. And hopefully there isn't a payment step partway through!
The fewer hard errors nix the whole workflow, the lower your ETTWS.
This is a thousand unusually smart monkeys who speak every major human language fluently and are proficient in every major programming language, but sometimes still make bizarre mistakes and need to be put back on track.
replyThis is fun for you?
replydeleted
reply