ZookeepergameCool173B to LocalLLaMA@poweruser.forumEnglish · 1 year agostrategies for training local models for low-resource languagesplus-squaremessage-squaremessage-square0fedilinkarrow-up11arrow-down10
arrow-up11arrow-down1message-squarestrategies for training local models for low-resource languagesplus-squareZookeepergameCool173B to LocalLLaMA@poweruser.forumEnglish · 1 year agomessage-square0fedilink
minus-squareZookeepergameCool173BtoLocalLLaMA@poweruser.forum•bellman-7b - a Swedish llama2 finetunelinkfedilinkEnglisharrow-up1·1 year agoTry to fine tune a 13b model instead, which has a way better command of Swedish than the 7B. And in my experience tends to have less issues with becoming repetitive etc. linkfedilink
Try to fine tune a 13b model instead, which has a way better command of Swedish than the 7B. And in my experience tends to have less issues with becoming repetitive etc.