schizoidman@lemm.ee to Technology@lemmy.worldEnglish · 2 个月前DeepSeek's distilled new R1 AI model can run on a single GPU | TechCrunchtechcrunch.comexternal-linkmessage-square22linkfedilinkarrow-up1177arrow-down117
arrow-up1160arrow-down1external-linkDeepSeek's distilled new R1 AI model can run on a single GPU | TechCrunchtechcrunch.comschizoidman@lemm.ee to Technology@lemmy.worldEnglish · 2 个月前message-square22linkfedilink
minus-squarefogetaboutit@programming.devlinkfedilinkEnglisharrow-up4arrow-down10·2 个月前ew probably still censored.
minus-squareT156@lemmy.worldlinkfedilinkEnglisharrow-up12arrow-down3·2 个月前The censorship only exists on the version they host, which is fair enough. If they’re running it themselves in China, they can’t just break the law. If you run it yourself, the censorship isn’t there.
minus-squarejaschen@lemm.eelinkfedilinkEnglisharrow-up4arrow-down1·2 个月前Untrue, I downloaded the vanilla version and it’s hardcoded in.
minus-squareMonkderVierte@lemmy.mllinkfedilinkEnglisharrow-up2arrow-down1·edit-22 个月前Yeah, i think the censoring in the LLM data itself would be pretty vulnerable to circumvention.
minus-squarefogetaboutit@programming.devlinkfedilinkEnglisharrow-up1·2 个月前if the model is censored… then what, retraining it? Or doing it from scratch like what open-r1 is doing?
minus-squarejaschen@lemm.eelinkfedilinkEnglisharrow-up2arrow-down3·2 个月前The self hosted model has hard coded censored content.
ew probably still censored.
The censorship only exists on the version they host, which is fair enough. If they’re running it themselves in China, they can’t just break the law.
If you run it yourself, the censorship isn’t there.
Untrue, I downloaded the vanilla version and it’s hardcoded in.
Yeah, i think the censoring in the LLM data itself would be pretty vulnerable to circumvention.
You can self host it right??
if the model is censored… then what, retraining it? Or doing it from scratch like what open-r1 is doing?
The self hosted model has hard coded censored content.