r-three/t-few

How long it will take for pretraining the model using A100(80G)?

Closed this issue · 2 comments

shuishen112 commented
How long it will take for pretraining the model using A100(80G)?
HaokunLiu commented

I roughly remember we rented a 4xA100 for half a week.

shuishen112 commented

got it. Thanks for your feedback.