Every time I hit a slump in (any) work, all I do is channel my inner world model of Rylan to get over any activation energy barriers and remind myself that you can, in fact, just do things
All I want to say while people are solving unsolved conjectures in their free time and tweeting them out
Here鈥檚 my claim in grad school slack from ~2 years ago right after completing a PhD deciding that I was done with academia after seeing the scale of things inside @nvidia
Replying to @_arohan_ @dheevatsa
Yep KL shampoo does not have this bug either. Stay tuned for a short paper with some runs!
Think you know SF well? People in SF interested in going for an SF themed geoguesser, here鈥檚 an invite to
SF GEOGUESSER CUP
Replying to @kalomaze
First pointed out by PSGD (@HessianFree )
arxiv.org/pdf/2402.11858
now everyone can do this
Someone has finally published nuances of how NCCL algorithms and protocols work.
Thank you so much to the authors since documentation is so scarce!
arxiv.org/abs/2507.04786










