It took Google 64 days to train PaLM using more than 6′000 TPU chips. Using the same setup (which is probably one of the most interconnected and capable ML training systems out there), it’d take 912 years.
It took Google 64 days to train PaLM using more than 6′000 TPU chips. Using the same setup (which is probably one of the most interconnected and capable ML training systems out there), it’d take 912 years.