Can you provide more explanation?
#4
by
Hugs4Llamas
- opened
Are you using the same 7b model twice? How can that have any other result than using it once?
Or are this random 2 out of the 8 experts of Dolphin Mixtral?
This is not the same model twice. This is 2 separate experts and is not affiliated with dolphin mixtral.
However, I have been experimenting with various architectures and will be updating this model soon. When I update I’ll include a more detailed write up about the models architecture.
Thanks for the answer
No problem, I have updated the model and included more information on the model card. Thanks for checking out the model.
macadeliccc
changed discussion status to
closed