Data2Behavior

zjunlp / Data2Behavior

Repository page

From Data to Behavior: Predicting Unintended Model Behaviors Before Training

4 0 Language: null License: MIT Updated: 10d ago

data2behavior large-language-model large-language-models natural-language-processing steering-behaviors steering-controls

README

We will release data and code for paper From Data to Behavior: Predicting Unintended Model Behaviors Before Training

0 AIs selected

Clear selection

Name

Task