[timzaman]

Iirc the bug Karpathy mentioned in his tweet was actually due to the seed being the same across multigpu data parallel workers! You need to account for this too. So the author hasnt solved it.

I know this bc I fixed the bug. And probably caused it. Hehe.

Also you dont just want to set ur numpy seed but also the native python one and the torch one.

[Salgat]

Yeah this isn't really a bug or even an issue with the library. If you instantiate an RNG with a seed and then fork your process, well duh of course the RNG will be repeated across the forks.