(Non-)literalness dataset of 1481 sentences formed with 184 Estonian particle verbs. Sentences are evaluated by 3 native speakers of Estonian on a 6-point scale [0,5] indicating the degree of compositionality of a particle verb. The first version of the dataset was introduced by Eleri Aedmaa, Maximilian Köper, Sabine Schulte im Walde (2018). Combining Abstractness and Language-specific Theoretical Indicators for Detecting Non-Literal Usage of Estonian Particle Verbs. In Proceedings of the NAACL 2018 Student Research Workshop (NAACL-SRW). New Orleans, LA.