NeuronScope provides a CLI and MCP server for mechanistic interpretability of language models. Built on TransformerLens, it helps researchers trace and visualize which neurons and attention heads drive specific model outputs. It is designed for AI researchers and engineers working on model transparency and debugging.
NeuronScope is a LLM eval & observability project. It focuses on understanding which specific neurons and attention heads are responsible for particular outputs in large language models. NeuronScope is an open-source project aimed at AI researchers. The project is open source (MIT). It runs on the command line.
RudrenduPaul builds and maintains NeuronScope, and it first shipped in 2026. Development happens publicly on GitHub with 8 commits in the last 90 days. Among its 4 catalogued features are Neuron Tracing, Attention Head Analysis, and TransformerLens Integration. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match