Base model
#2
by
ehartford
- opened
Is this based on Experiment26?
@ehartford
not directly, however, Experiment26 was used at the very beginning of the process in 0.1 when I did a SFT, then got DPO to 0.1.1, then merging starts from 0.2 all the way to 0.9 either among themselves or with other top 7B models.