How flux.1 dev Guiandec Embeding working ?
#3
by
GuardSkill
- opened
Great job, but I even don't know the theory of flux.1' Guiandec Embeding.
GuardSkill
changed discussion title from
Why reduce the parameters from 12B to 8B, for better trainning?
to How flux.1 dev Guiandec Embeding working ?