Jesus@lemmy.world to Political Memes@lemmy.world · 7 days agoWhat could possibly go wronglemmy.worldimagemessage-square68fedilinkarrow-up1505arrow-down146
arrow-up1459arrow-down1imageWhat could possibly go wronglemmy.worldJesus@lemmy.world to Political Memes@lemmy.world · 7 days agomessage-square68fedilink
minus-squarecybervseas@lemmy.worldlinkfedilinkEnglisharrow-up53arrow-down2·7 days agoIt’s open source. Apparently folks have already made mods of it that add CCP-sensitive info back in. Disclaimer: I have yet to see this for myself.
minus-squareperviouslyiner@lemmy.worldlinkfedilinkarrow-up4·edit-26 days agojust running it locally, apparently. The output of this model is being filtered by another AI, but only on the public-hosted copy.
minus-squareEven_Adder@lemmy.dbzer0.comlinkfedilinkEnglisharrow-up50arrow-down2·7 days agoThe answer I got out of DeepSeek-R1-Distill-Llama-8B-abliterate.i1-Q4_K_S
minus-squaretaiyang@lemmy.worldlinkfedilinkarrow-up27arrow-down1·7 days agoSo a real answer, basically. Too bad your average person isn’t going to bother with that. Still nice it’s open source.
minus-squarefelixwhynot@lemmy.worldlinkfedilinkarrow-up10·7 days agoSeems like the model you mentioned is more like a fine tuned Llama? Specifically, these are fine-tuned versions of Qwen and Llama, on a dataset of 800k samples generated by DeepSeek R1. https://github.com/Emericen/deepseek-r1-distilled
minus-squareEven_Adder@lemmy.dbzer0.comlinkfedilinkEnglisharrow-up8·edit-27 days agoYeah, it’s distilled from deepseek and abliterated. The non-abliterated ones give you the same responses as Deepseek R1.
minus-squarestebo@lemmy.dbzer0.comlinkfedilinkarrow-up3arrow-down1·7 days agoif it’s open source, can we also see what words/topics are being blocked?
It’s open source. Apparently folks have already made mods of it that add CCP-sensitive info back in. Disclaimer: I have yet to see this for myself.
just running it locally, apparently. The output of this model is being filtered by another AI, but only on the public-hosted copy.
The answer I got out of DeepSeek-R1-Distill-Llama-8B-abliterate.i1-Q4_K_S
So a real answer, basically. Too bad your average person isn’t going to bother with that. Still nice it’s open source.
Seems like the model you mentioned is more like a fine tuned Llama?
https://github.com/Emericen/deepseek-r1-distilled
Yeah, it’s distilled from deepseek and abliterated. The non-abliterated ones give you the same responses as Deepseek R1.
deleted by creator
if it’s open source, can we also see what words/topics are being blocked?