Hi everybody!
Inspired by a recent thread, mentioning the insane goliath abilities I decided to merge four SFT Yi models to make 2 seperate 55B Yi, one with context 200K and one with 32K.
Try them out and let me know!
Hi everybody!
Inspired by a recent thread, mentioning the insane goliath abilities I decided to merge four SFT Yi models to make 2 seperate 55B Yi, one with context 200K and one with 32K.
Try them out and let me know!
What are the eval results?