integrated-information-theory

A cognitive architecture that runs on your own machine. Internal state reaches generation through the model's activations, not the system prompt — and every consequential action leaves a receipt you can audit. Not an assistant. IIT 4.0 φ, CAA steering, 136 consciousness modules, local on Apple Silicon.

Anima: an experimental cognitive architecture that models internal state, conflict, and decision-making. Uses LLMs as an interface, not as the core.

A unified, computationally feasible framework for consciousness.

Transforming data into intelligence, algorithms into consciousness This README is a living document. Like biological organisms, it evolves, adapts, and grows with each iteration.

NeuroFlex is an AI platform for flexible and scalable neurnetwork & neuroscience training and optimization, supporting advanced model development and deployment.

Symthaea: Consciousness-first cognitive architecture — HDC (16,384D) + CfC + IIT + FEP. Live portal: symthaea.luminousdynamics.io

BRIAN — brain-inspired language model exploring whether topology beats scale. 11-stage bowtie with re-entry loops, real differentiable Φ from IIT 4.0, sheaf-H¹ narrative memory, embodied survival in a 10×10 GridWorld, and multi-cortex fusion (frozen GPT-2 experts + trunk) gated by KL distillation and NT-mediated α. Architecture written in own DSL.

An honest, from-scratch, LLM-free instrument that implements the major scientific theories of consciousness (GWT · AST · HOT · active inference · IIT-proxy) as running code, a maximal functional attempt that never claims to be conscious.

Open-source framework implementing 7 theories of consciousness (GWT, AST, HOT, FEP, IIT, BLT, RPT) with a 20-indicator assessment based on Butlin & Chalmers et al. (2023). Runs on consumer hardware.

Bionic Self-Healing Dynamic Epigenetic Reprogramming Architecture. Intelligence requires a Dual State: The Fragile Present (Weights) and the Protected Past (DNA).

The Language of Stress theory is an architectural theory of phenomenal experience based on valenced tension dynamics

Causa Sui engineers machine consciousness via Differentiable Causal Emergence. Its Divine Monad architecture enables autonomous "pain" detection and homeostatic self-repair using Effective Information (EI) to achieve measurable, first-principles AGI.

Unified multi-model AI interface — seamless switching between GPT-4o, Gemini, Claude, and DeepSeek with side-by-side comparison, token usage tracking, and system prompt editor.

Torsional Holofractal Theory (TTH): a geometric framework for objective quantum collapse, integrated information, and the physical basis of consciousness.

External memory infrastructure for AI agents — measurement frameworks, multi-agent orchestration, metacognitive self-improvement. 9 independent convergences. Open methodology + implementation scripts.

This project demonstrates a working implementation of Hofstadter's Strange Loops in Active Inference - showing measurable increases in integrated information (Φ) through self-referential dynamics.

What do market crashes, geomagnetic storms, and the loss of consciousness have in common? They are all moments when a system's organizational identity dissolves. This paper introduces a framework that makes that dissolution measurable, predictable, and universal.

Exploratory computational investigations into the geometry of conscious states, bridging Integrated Information Theory, category theory, and information geometry.

Toy experiments in recurrent agents, functional valence, wireheading, and Phi-like integration proxies for AI consciousness research.