minus-squarejxjq@alien.topBtoLocalLLaMA•Could multiple 7b models outperform 70b models?linkfedilinkEnglisharrow-up1·1 year agoThank you for sharing, I understand now linkfedilink
minus-squarejxjq@alien.topBtoLocalLLaMA•Could multiple 7b models outperform 70b models?linkfedilinkEnglisharrow-up1·1 year agoDoes this use of mixture-of-experts mean that multiple 70b models would perform ?better than multiple 7b models linkfedilink
Thank you for sharing, I understand now