inari@piefed.zip to Technology@lemmy.worldEnglish · edit-28 days agoI gave Qwen 3.8 27B a reverse-engineering job I assumed needed a frontier model, and it finished in 30 minuteswww.xda-developers.comexternal-linkmessage-square34fedilinkarrow-up1137arrow-down133 cross-posted to: technology@lemmy.ziptechnology@lemmy.mlhackernews@lemmy.bestiver.se
arrow-up1104arrow-down1external-linkI gave Qwen 3.8 27B a reverse-engineering job I assumed needed a frontier model, and it finished in 30 minuteswww.xda-developers.cominari@piefed.zip to Technology@lemmy.worldEnglish · edit-28 days agomessage-square34fedilink cross-posted to: technology@lemmy.ziptechnology@lemmy.mlhackernews@lemmy.bestiver.se
minus-squarefrongt@lemmy.ziplinkfedilinkEnglisharrow-up1·7 days agoThe solution is to train a LoRA. It takes the base model, and overlays all your data on top.
minus-squarehirihit640@sh.itjust.workslinkfedilinkEnglisharrow-up1·7 days agoThat sounds like a lot of work to just read some docs. RAG doesn’t require any training at all. I assume there are other approaches as well that would be faster than LoRA
The solution is to train a LoRA. It takes the base model, and overlays all your data on top.
That sounds like a lot of work to just read some docs. RAG doesn’t require any training at all. I assume there are other approaches as well that would be faster than LoRA