logoalt Hacker News

HarHarVeryFunny • today at 5:30 PM • 2 replies • view on HN

Any idea why Alibaba might ship a model while aware that it has this defect when it is so easy to fix?

Is this looping tough to fix by changing the training?


Replies

bhu8 • today at 5:48 PM

Steering/model weight lobotomy usually have a ton of side effects. You fix one defect and add 10 others.