Why is the model size in Evo2 set to 40.3 B parameters? Is it because of the 9.3 T tokens of training data?
#2
by RevengeUSA - opened
Why is the model size in Evo2 set to 40.3 B parameters? Is it because of the 9.3 T tokens of training data?
Why is the model size in Evo2 set to 40.3 B parameters? Is it because of the 9.3 T tokens of training data?