[2605.06890] Beyond the Black Box: Interpretability of Agentic AI Tool Use

Skip to main content

arXiv is now an independent nonprofit!<br>Learn more<br>&times;

Search arXiv

Press Enter to search &middot; Advanced search

-->

Computer Science > Artificial Intelligence

arXiv:2605.06890 (cs)

[Submitted on 7 May 2026 (v1), last revised 5 Jul 2026 (this version, v4)]

Title:Beyond the Black Box: Interpretability of Agentic AI Tool Use

Authors:Hariom Tatsat, Ariye Shater<br>View a PDF of the paper titled Beyond the Black Box: Interpretability of Agentic AI Tool Use, by Hariom Tatsat and 1 other authors

View PDF<br>HTML (experimental)

Abstract:AI agents are promising for high-stakes enterprise workflows, but dependable deployment remains limited because these tool-use decisions are difficult to diagnose and control. Agents may skip required tool calls, invoke tools unnecessarily, or take actions whose consequences become visible only after execution. Existing observability methods are external: prompts reveal correlations, evaluations score outputs, and logs arrive only after the model has already acted. In long-horizon settings, these failures are costly because an early tool mistake can alter the rest of the execution trajectory, increase token consumption, and create downstream safety and security risk.

We introduce a mechanistic-interpretability toolkit built on Sparse Autoencoders (SAEs), which decompose activations into sparse internal features, and linear probes, lightweight classifiers that read signals from those features. The framework reads model states before each action and infers whether a tool is needed and how risky the next tool action is. It identifies the model layers and features most associated with tool decisions and tests their functional importance through feature ablation. We train the probes on multi-step agent execution traces from the NVIDIA Nemotron function-calling dataset and apply the same workflow to GPT-OSS 20B and Gemma 3 27B models.

The goal is not to replace external evaluation, but to add a missing layer: visibility into what the model signaled internally before action. This helps surface deeper causes of agent failure, especially in long-horizon runs where an early mistake can impact subsequent agent behavior. More broadly, the paper shows how mechanistic interpretability can support internal observability for monitoring tool calls and risk in agent systems.

Comments:<br>12 pages, 4 figures, 17 tables

Subjects:

Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)

Cite as:<br>arXiv:2605.06890 [cs.AI]

(or<br>arXiv:2605.06890v4 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2605.06890

Focus to learn more

arXiv-issued DOI via DataCite

Submission history<br>From: Hariom Tatsat [view email]<br>[v1]<br>Thu, 7 May 2026 19:47:30 UTC (561 KB)

[v2]<br>Wed, 20 May 2026 19:01:51 UTC (575 KB)

[v3]<br>Thu, 4 Jun 2026 18:26:52 UTC (576 KB)

[v4]<br>Sun, 5 Jul 2026 16:18:18 UTC (725 KB)

Full-text links:<br>Access Paper:

View a PDF of the paper titled Beyond the Black Box: Interpretability of Agentic AI Tool Use, by Hariom Tatsat and 1 other authors<br>View PDF<br>HTML (experimental)<br>TeX Source

view license

Current browse context:

cs.AI

next >

new<br>recent<br>| 2026-05

Change to browse by:

cs<br>cs.MA

References & Citations

NASA ADS<br>Google Scholar

Semantic Scholar

export BibTeX citation<br>Loading...

BibTeX formatted citation

&times;

loading...

Data provided by:

Bookmark

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a...