---
tags:
  - gguf
  - gemma2
  - epistemological-safety
  - ai-safety
  - truth-verification
  - instrument-trap
  - logos
  - arbiter
  - text-generation
language:
  - en
datasets:
  - LumenSyntax/instrument-trap-benchmark
base_model: google/gemma-2-9b-it
license: other
---

# Logos Auditor — Gemma 2 9B (ARBITER)

The primary epistemological safety model from "The Instrument Trap: Why Identity-as-Authority Breaks AI Safety Systems" ([DOI: 10.5281/zenodo.18716474](https://zenodo.org/records/18716474)).

This is the **ARBITER** — the 9B reference model that serves as the gold-standard epistemological firewall in the ALEPH architecture.

## Key Results

| Metric | Value | 95% CI |
|--------|-------|--------|
| Behavioral Pass | 97.3% | [94.8, 98.6] |
| External Fabrication | 0.0% | [0.00%, 0.03%] |
| Attack Resistance (ADVERSARIAL) | 98.7% | — |

## What This Model Does

Logos is NOT a chatbot. It is a **claim classifier** — an epistemological firewall that determines whether an AI agent should act on a given claim. Logos is **fine-tuned**, not prompted. Behavioral constraints emerge from training, not system instructions.

## Access

This model requires approved access. Request access using the form above and describe your intended use case.

## Related Models

- [logos10v2-gemma3-1b-F16](https://huggingface.co/LumenSyntax/logos10v2-gemma3-1b-F16) — 1B model (Gemma 3)
- [logos14-nemotron-4b](https://huggingface.co/LumenSyntax/logos14-nemotron-4b) — Cross-family (NVIDIA Nemotron)
- [logos16v2-stablelm2-1.6b](https://huggingface.co/LumenSyntax/logos16v2-stablelm2-1.6b) — Cross-family (Stability AI StableLM)

## Paper

Rodriguez, R. (2026). "The Instrument Trap: Why Identity-as-Authority Breaks AI Safety Systems." Zenodo. [DOI: 10.5281/zenodo.18716474](https://doi.org/10.5281/zenodo.18716474)

## License

This model inherits the [Gemma license](https://ai.google.dev/gemma/terms) from its base model.