With the surge of the application of machine learning (ML) systems in our daily life there is an increasing demand to make operation and results of these systems interpretable for people with different backgrounds (ML experts, non-technical experts etc.). A wide range of research exists, particular in ML research on specific interpretability techniques (e.g., extracting and displaying information from ML pipelines). However, often a background in machine learning or mathematics is required to interpret the results of the interpretability technique itself. Therefore there is an urgent lack of techniques which may help non-technical experts in using such systems. The grounding hypothesis of this analysis is that, especially for non-technical experts, context is an influential factor in how people make sense of complex algorithmic systems. Therefore an interaction between a user and an application assumed to be an interplay between a user and his historical context, the context of the situation in which the interaction is embedded and the algorithmic system. Interpretability techniques are the common link which bring all these different aspects together. In order to evaluate the assumption that most of the current interpretability research is tailored to a technical audience and gain an overview over existing interpretability techniques we conducted a literature mapping study studying the state of interpretability research in the field of natural language processing (NLP). The results of this analysis suggest that indeed most techniques are not evaluated in a context where a non-technical expert may use it and that even most publications lack a proper definition of interpretability. Keywords: Literature Mapping Study, Interpretability Research, Natural Language Processing.