Gemma 2B with 10M context length

Posted by mustafaaljadery on Thursday May 09, 2024 @02:14PM
from the github.com dept.

( 1 comment | 6 points | on HN )

Re:Gemma 2B with 10M context length by FloatArtifact on Thursday May 09, 2024 @04:25PM (#40312606)

The question is how such a model performs proportionate to token utilization.

[ Parent | on HN ]