eDIF: A European Deep Inference Fabric for Remote Interpretability of LLM

TL;DR

This paper introduces eDIF, a European infrastructure for remote interpretability of large language models, demonstrating its technical feasibility, usability, and potential to democratize access for researchers across Europe.

Contribution

The paper presents the design, deployment, and evaluation of eDIF, a novel European infrastructure enabling remote LLM interpretability research with a focus on accessibility and community building.

Findings

Stable platform performance during pilot study

Positive user engagement and feedback

Identified limitations with data download and execution interruptions

Abstract

This paper presents a feasibility study on the deployment of a European Deep Inference Fabric (eDIF), an NDIF-compatible infrastructure designed to support mechanistic interpretability research on large language models. The need for widespread accessibility of LLM interpretability infrastructure in Europe drives this initiative to democratize advanced model analysis capabilities for the research community. The project introduces a GPU-based cluster hosted at Ansbach University of Applied Sciences and interconnected with partner institutions, enabling remote model inspection via the NNsight API. A structured pilot study involving 16 researchers from across Europe evaluated the platform's technical performance, usability, and scientific utility. Users conducted interventions such as activation patching, causal tracing, and representation analysis on models including GPT-2 and DeepSeek-R1-70B. The study revealed a gradual increase in user engagement, stable platform performance throughout, and a positive reception of the remote experimentation capabilities. It also marked the starting point for building a user community around the platform. Identified limitations such as prolonged download durations for activation data as well as intermittent execution interruptions are addressed in the roadmap for future development. This initiative marks a significant step towards widespread accessibility of LLM interpretability infrastructure in Europe and lays the groundwork for broader deployment, expanded tooling, and sustained community collaboration in mechanistic interpretability research.