This repository contains code for NAACL'2024 findings Paper "Do Prompt Positions Really Matter?".
We use Language Model Evaluation Harness framework (https://github.com/EleutherAI/lm-evaluation-harness) with slight modifications (from late 2023 in lm-eval-harness folder).