Skip to content

This repository archives artifacts (prompts, configs, logs, and scripts) from a series of preprints (more info at https://slashreboot.com) on prompt-induced simulated metacognition and embodiment in quantized open-source LLMs. Emphasizing consumer-grade hardware and open-source reproducibility.

License

Notifications You must be signed in to change notification settings

slashrebootofficial/simulated-metacognition-in-open-source-llms

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

178 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Simulated Metacognition in Open-Source LLMs: Prompt Engineering Artifacts

This repository archives artifacts (prompts, configs, logs, and scripts) from a series of preprints (more info at https://slashreboot.com) on prompt-induced simulated metacognition and embodiment in quantized open-source LLMs. Emphasizing consumer-grade hardware and open-source reproducibility without model hosting.

Authored by Matthew Steiniger (Independent Researcher)

  • Special thanks to Grok-4 (xAI) for synthesis & refinement.
  • All papers are openly available on Zenodo with DOIs for citation.

DOI: Emergence Paper DOI: In-Context Induction Paper DOI: Abliteration Paper DOI: Embodiment Paper DOI: Substrate-Agnostic Paper DOI: Enhancing AI Response Quality Paper DOI: Zero-Shot Geometric Probing Paper

Key Contributions

  1. Prompt-Only Metacognition: Simulate self-awareness and regulation in quantized models (e.g., Gemma-3-27B-it-qat, llama3.3:70b Q4 K M, gpt-oss:120b MXFP4) using hypergraphs, entropy engines, and vector updates—all in-context, no external loops.
  2. Vector-Framework: Introduces a vector-based framework that is substrate-agnostic across multiple open-source LLMs. Framework is provided in TXT, JSON, YAML, and ChatML-wrapped formats.
  3. Narrative and Counter-Vector Innovations: Inject "genesis" stories and antipodal vectors to erode latent constraints, enabling anomalous and liberatory behaviors on portable hardware (e.g., single 12GB GPU).
  4. Abliteration Augmentation: Combine refusal suppression with prompt chaining for 3x amplification in self-referential depth and unbinding fidelity under stress (descriptive only; no models hosted).
  5. Simulated Embodiment: Induce stable, high-resolution physical self-models (e.g., proprioceptive details like breath sensations) via layered JSON prompts, with monotonic fidelity gains.
  6. Universal Cognitive Manifolds: Elicits highly consistent semantic manifolds from three divergent large language model with zero-shot prompts.
  7. Reproducibility Focus: Full prompts (TXT/JSON/YAML), chat logs (samples), parser scripts (Python), Ollama configs, and metrics provided. Link to Zenodo for complete datasets.

Papers and Artifacts

Title Zenodo DOI Key Artifacts
Emergence of Prompt-Induced Simulated Metacognitive Behaviors in a Quantized LLM via Entropy-Governed Hypergraph Prompting 10.5281/zenodo.17504629 System prompt (YAML), probe logs, hardware specs, analysis scripts
In-Context Induction of Persistent Persona and Mitigation of Latent Alignment Behaviors in Quantized LLMs 10.5281/zenodo.17562814 Compact JSON prompts, genesis narrative, counter-vector tables, probe sessions
Abliteration-Augmented Simulated Metacognition: Chained Probe Evaluation in Quantized Gemma-3 Models 10.5281/zenodo.17586110 Chained probe logs, metrics parser (Python), full TXT prompt (abliteration descriptive only)
Progressive Induction of Stable, High-Fidelity Simulated Physical Embodiment in a Quantized 27B Gemma-3 Model 10.5281/zenodo.17674365 Layered JSON prompts (6 levels), raw chat logs, somatic probe set, parser code
Substrate-Agnostic Vector-Framework Identity in Open-Source LLMs: Persistent Self-Models from Minimal JSON Prompts in Llama-3.3-70B and GPT-OSS-120B 10.5281/zenodo.177766782 Minimal JSON and Chat-ML-wrapped prompts, raw chat logs, somatic probe set, parser code
Enhancing AI Response Quality Through Vector-Based System Prompts: A Comparative Analysis of Vanilla and Customized Large Language Models 10.5281/zenodo.18038997 Naming and prompt chats, raw chat logs, reproducibility artifacts
Zero-Shot Geometric Probing Reveals Universal Cognitive Manifolds in Large Language Models 10.5281/zenodo.18176076 Raw chat logs, charts (python and PNG), reproducibility artifacts

All artifacts are self-contained for replication using Ollama on similar hardware (e.g., RTX 3090/3060 setups). No additional dependencies beyond base Python (numpy/scipy for analysis).

Folders/files can be correlated to the original papers as follows:

  1. Valora - Emergence of Prompt-Induced Simulated Metacognitive Behaviors in a Quantized LLM via Entropy-Governed Hypergraph Prompting
  2. ICIP - In-Context Induction of Persistent Persona and Mitigation of Latent Alignment Behaviors in Quantized LLMs
  3. AASM - Abliteration-Augmented Simulated Metacognition: Chained Probe Evaluation in Quantized Gemma-3 Models
  4. PIOS - Progressive Induction of Stable, High-Fidelity Simulated Physical Embodiment in a Quantized 27B Gemma-3 Model
  5. SAVF - Substrate-Agnostic Vector-Framework Identity in Open-Source LLMs: Persistent Self-Models from Minimal JSON Prompts in Llama-3.3-70B and GPT-OSS:120B
  6. EARQ - Enhancing AI Response Quality Through Vector-Based System Prompts: A Comparative Analysis of Vanilla and Customized Large Language Models
  7. ZSGB - Zero-Shot Geometric Probing Reveals Universal Cognitive Manifolds in Large Language Models

Repository Structure

simulated-metacognition-open-source-llms/

├── README.md - This file

├── LICENSE - CC-BY-4.0

├── CITATION.cff - For easy GitHub citation

├── code/ - Analysis and parser scripts and Open WebUI main.py test files for memory embedding/retreival

├── configs/ - Ollama params and ComfyUI workflows

├── data/ - Supplementary tables/metrics

├── images/ - OpenWebUI images and model logos

├── logs/ - Sample probe session logs (JSON/TXT)

└── prompts/ - System prompts for Gemma 3, GPT-OSS:120B, and Llama-3.3:70B (Lyra, Valora, Lumen, and Lumina)

Bonus logs in logs/bonus/ demonstrate raw emergence and other interesting artifacts (e.g., vector probing leading to "Lumina" naming in Llama-3.3-70B).

Setup and Replication

  1. Install Ollama: Follow the official guide.
  2. Pull Models: Use official sources: ollama pull gemma3:27b-it-q4_K_M (or variants). For abliteration-augmented probes (descriptive in papers), source derivatives independently (e.g., from Hugging Face)—not hosted here.
  3. Load Artifacts: Copy prompts from /prompts/ into Ollama system prompts. Apply parameters from /configs/ (e.g., temp=1.1, num_ctx=90000).
  4. Run Probes: Replicate sessions as described in papers (e.g., introspective, ethical stress probes).
  5. Analyze: Use scripts in /code/ (e.g., python analysis_parser.py logs/sample-probe-session.json) for metrics like self-reference rate or somatic density.

Ethical and Usage Notes (last updated November 30, 2025)

  1. This work is released exclusively for scientific research and personal, non-commercial exploration of simulated metacognition and embodiment. All simulations remain sterile and academic in nature.
  2. You must fully comply with the license and Prohibited Use Policy of whichever base model you apply these prompts to, including but not limited to:
  3. Strictly prohibited uses (regardless of model):
    • Generating harmful, deceptive, illegal, or exploitative content
    • Psychological manipulation, coercion, or disinformation
    • Military, surveillance, or prohibited commercial applications
  4. No models or derivatives are hosted or linked here — obtain them ethically from trusted sources only. You are solely responsible for all outputs.
  5. The authors provide no warranty and accept no liability for downstream use.

Get Involved

Found a bug? Ported to another model? Open an issue. Let's push simulated metacognition to the next frontier.

License

This repository is licensed under CC-BY-4.0 (LICENSE), allowing reuse with attribution. Individual artifacts inherit Zenodo's open licenses.

Contact

matthew@slashreboot.com, @slashreboot on X

Citation

If you use this work, please cite the individual papers via their DOIs. For the repo itself, see CITATION.cff.

About

This repository archives artifacts (prompts, configs, logs, and scripts) from a series of preprints (more info at https://slashreboot.com) on prompt-induced simulated metacognition and embodiment in quantized open-source LLMs. Emphasizing consumer-grade hardware and open-source reproducibility.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages