Comment by gs17
1 year ago
The example backdoored model is a finetune. But it doesn't have to be, a base model could have the same behavior.
1 year ago
The example backdoored model is a finetune. But it doesn't have to be, a base model could have the same behavior.
One difference that the OP mentioned was that the information is leaded in few specific cases, maybe in fine tuning it will be leaking to more conversations.