Re:Open-R1: an open reproduction of DeepSeek-R1
by anthk on Tuesday January 28, 2025 @08:01AM
(#42851817)
And it will create horrible un-debuggable bugs, human-killing causalities and who knows what more. Welcome to Idiocracy.
[ Parent
| on HN ]
Re:Open-R1: an open reproduction of DeepSeek-R1
by __MatrixMan__ on Tuesday January 28, 2025 @01:15PM
(#42855810)
So far I've found the bugs that it writes to be indistinguishable from the bugs that I'd write. Like, you look at it and think:
> Oh I see what you were going for, but no...
It's learning its deficiencies from us. That's concerning for many reasons, but "un-dubuggable bugs" is pretty far down on the list for me.