You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When pretraining, the paper talked about packing data.
If so, I think all input data will be 512 tokens.
In the currently implemented code, the masking strategy is deterministic.
So it is assumed that there will always be the same number of masking for 512 tokens.
is this right?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
When pretraining, the paper talked about packing data.
If so, I think all input data will be 512 tokens.
In the currently implemented code, the masking strategy is deterministic.
So it is assumed that there will always be the same number of masking for 512 tokens.
is this right?
Beta Was this translation helpful? Give feedback.
All reactions