Experimental merges
Collection
1 item
•
Updated
This is a merge of pre-trained language models
Goal of this merge was to explore limits of 12b merging
This merge is surprisingly stable, it's logical capapabilities are ok, already good prose of Badman and Unity are enriched by some layers of Delta-Vector/Rei-12B.
It still need some more testing.
Use ChatML
Tested on T 1.01