minus-squareACbHrhMJ@lemmy.worldtoNo Stupid Questions@lemmy.world•Do LLM modelers maintain a list of manual corrections fed by humans?linkfedilinkarrow-up3arrow-down1·13 days agoIf the model does something undesirable or wrong, it is given the equivalent of a shock with a cattle prod. With repetition, this process reshapes the network and the model avoids the ‘bad’ areas. linkfedilink
minus-squareACbHrhMJ@lemmy.worldtoMildly Infuriating@lemmy.world•Inconsiderate fucks who litterlinkfedilinkEnglisharrow-up8·1 month agoYou could always collect it all and dump it on the guy’s doorstep? linkfedilink
minus-squareACbHrhMJ@lemmy.worldtoNo Stupid Questions@lemmy.world•*Permanently Deleted*linkfedilinkarrow-up2arrow-down2·2 months agoAt least give it a while before you try linkfedilink
If the model does something undesirable or wrong, it is given the equivalent of a shock with a cattle prod. With repetition, this process reshapes the network and the model avoids the ‘bad’ areas.