2024-12-17
Torch code bug that took me an hour to fix: I wrapped a function in
@torch.inference_mode which called another function which
called another function that was trying to call
torch.backward.
Booking this site fonts for mathematics for later use.
Intrigued by the idea of building my own Bluesky feed using Graze. Maybe some day I will have time to.
My colleague from USC pointed me to their incomplete paper about training dynamics that they submitted to ICLR. I mean to give it a skim sometime.