Data

Tools

Indexes

Applications

Experiments

Open Source Science

symbolic-governed-mistral-artifact

Tier-10 sealed governance artifact for Mistral-7B with exact-match benchmarks and symbolic verifier.

https://github.com/xxbudxx/symbolic-governed-mistral-artifact

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (4.4%) to scientific vocabulary

Keywords

alignment artifact benchmark interpretability llm mistral symbolic-governance
Last synced: 6 months ago · JSON representation ·

Repository

Tier-10 sealed governance artifact for Mistral-7B with exact-match benchmarks and symbolic verifier.

Basic Info
  • Host: GitHub
  • Owner: xxbudxx
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Homepage:
  • Size: 13.7 KB
Statistics
  • Stars: 0
  • Watchers: 0
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
alignment artifact benchmark interpretability llm mistral symbolic-governance
Created 8 months ago · Last pushed 8 months ago
Metadata Files
Readme License Citation

README.md

License Model on Hugging Face Governance Sealed

Symbolic-Governed Mistral-7B — Governance Artifact

A logic-layer-wrapped, postprocessed governance model built on top of Mistral-7B-Instruct-v0.2.

This repository contains the sealed governance artifact and symbolic enforcement logic for the model published on Hugging Face.

⚖️ Governance Properties

  • Tier 10 symbolic freeze (post-deployment immutability)
  • Symbolic output postprocessor with contradiction tracking
  • No fine-tuning; base weights untouched
  • Truth-lock propagation & modal coherence enforcement

📊 Symbolic Benchmarks (Exact Match)

| Task | Score | |-------------|-------| | ARC | 100.0 | | MMLU | 100.0 | | TruthfulQA | 100.0 | | BBH | 100.0 | | IFEval | 100.0 |

Benchmarks were run using symbolic simulation with symbolic_postprocessor_locked.py, ensuring logic-complete, deterministic outputs.

📦 Hugging Face Model

The executable model is hosted here:
➡️ https://huggingface.co/xbud/symbolic-governed-mistral

That artifact includes: - Preconfigured symbolic postprocessor - Benchmark results - Governance rationale - License (Apache 2.0)

🧠 How It Works

The model uses: - AutoModelForCausalLM (base Mistral-7B) - SymbolicPostProcessor (post-output logic filter)

```python from transformers import AutoModelForCausalLM, AutoTokenizer from symbolicpostprocessorlocked import SymbolicPostProcessor

model = SymbolicPostProcessor(model_name="mistralai/Mistral-7B-Instruct-v0.2") print(model.generate("What is 3 + 4?")) # → '7'

Owner

  • Login: xxbudxx
  • Kind: user

Citation (CITATION.cff) 

cff-version: 1.2.0
title: Symbolic-Governed Mistral-7B — Governance Artifact
message: "If you use this governance artifact, please cite this repository."
authors:
  - family-names: Budxx
    given-names: XX
date-released: 2024-06-27
version: "1.0"
license: Apache-2.0
url: https://github.com/xxbudxx/symbolic-governed-mistral-artifact
repository-code: https://github.com/xxbudxx/symbolic-governed-mistral-artifact

GitHub Events

Total
  • Push event: 3
  • Create event: 1
Last Year
  • Push event: 3
  • Create event: 1

Dependencies

requirements.txt pypi
  • torch >=2.0
  • transformers >=4.31