🃏Joker@sh.itjust.works to Comic Strips@lemmy.worldEnglish · edit-21 month ago[Kevin Comics] Zuckerberg AIsh.itjust.worksexternal-linkmessage-square20fedilinkarrow-up1768arrow-down115file-text
arrow-up1753arrow-down1external-link[Kevin Comics] Zuckerberg AIsh.itjust.works🃏Joker@sh.itjust.works to Comic Strips@lemmy.worldEnglish · edit-21 month agomessage-square20fedilinkfile-text
minus-squareryedaft@sh.itjust.workslinkfedilinkEnglisharrow-up1·1 month agoHow on earth would you distribute the model for inference without the weights? The gradients are obviously gone so you can’t continue training on the model. Maybe you can still do some kind of LORA?
![[Kevin Comics] Zuckerberg AI](https://sh.itjust.works/pictrs/image/2896b29d-18c5-4302-89fa-c5c9cb8b8aa7.png){kind=link}
How on earth would you distribute the model for inference without the weights? The gradients are obviously gone so you can’t continue training on the model. Maybe you can still do some kind of LORA?