Measuring Harmful Representations in Scandinavian Language Models

Abstract: Scandinavian countries are perceived as role-models when it comes to gender equality. With the advent of pre-trained LLMs and their widespread usage, we investigate to what extent gender-based harmful and toxic content exist in selected Scandinavian LLMs. We examine nine models, covering Danish, Swedish, and Norwegian, by manually creating template-based sentences and probing the models for completion. We evaluate the completions using two methods for measuring harmful and toxic completions and provide a thorough analysis of the results. We show that Scandinavian pre-trained LLMs contain harmful and gender-based stereotypes with similar values across all languages. This finding goes against the general expectations related to gender equality in Scandinavian countries and shows the possible problematic outcomes of using such models in real-world settings.