Skip to main content

Felix Morger

I'm interested in the evaluation and interpretability of large language models (LLMs). To this end, some of my research interests have been

  • Comparing eye-tracking data and importance values of LLMs to see if humans and language models process focus on the same words (Morger et. al, 2022).
  • English-Swedish language transfer using natural language inference (NLI) (Morger, 2023).
  • Creating a Swedish benchmark (SuperLIM) for natural language understanding.
  • Using large language models to study linguistic variation. In particular, I'm investigating whether large language models can predict the omission or inclusion of "att" in the "kommer (att)"-construction in Swedish.

Publications BibTeX

2023 BibTeX

2022 BibTeX

2020 BibTeX

Profile picture for user


  • PhD student
  • Research engineer

Blog entries and news