🃏Joker@sh.itjust.works to Technology@lemmy.worldEnglish · 22 hours agoMake illegally trained LLMs public domain as punishmentwww.theregister.comexternal-linkmessage-square159fedilinkarrow-up11.36Kfile-text
arrow-up11.36Kexternal-linkMake illegally trained LLMs public domain as punishmentwww.theregister.com🃏Joker@sh.itjust.works to Technology@lemmy.worldEnglish · 22 hours agomessage-square159fedilinkfile-text
minus-squareFaceDeer@fedia.iolinkfedilinkarrow-up1·9 hours agoI’ve been working with local LLMs for over a year now. No guardrails, and many of them fine-tuned against censorship. They can’t output arbitrary training material verbatim. Llama 3 was trained on 15 trillion tokens, both the 8B and 70B parameter versions.. So around 1:1000, not 1:7.
I’ve been working with local LLMs for over a year now. No guardrails, and many of them fine-tuned against censorship. They can’t output arbitrary training material verbatim.
Llama 3 was trained on 15 trillion tokens, both the 8B and 70B parameter versions.. So around 1:1000, not 1:7.