Uploaded finetuned model

• Developed by: TesseractLens-AI
• License: apache-2.0
• Finetuned from model : unsloth/Qwen2.5-1.5B-Instruct-unsloth-bnb-4bit

This qwen2 model was trained 2x faster with Unsloth

---

license: apache-2.0 language:

• en tags:
• unsloth
• qwen2.5
• spatial-reasoning
• tesseract-lens
• ai-safety
• deterministic-alignment

---

Tesseract-Qwen2.5-1.5B-Spatial

The Tesseract Lens Engine — High-Dimensional Spatial Reasoning & Deterministic AI Safety Architecture

🌌 Executive Overview

Tesseract-Qwen2.5-1.5B-Spatial is a specialized fine-tuned open-weights model built on Qwen-2.5-1.5B-Instruct using Unsloth LoRA adapters ($r=16$).

Developed by TesseractLens-AI, this model bridges the gap between probabilistic neural networks and deterministic symbolic logic. By projecting multi-variable logic rules and spatial state transformations through high-dimensional hypercube coordinate mappings, this architecture moves beyond traditional post-hoc text filters toward structural, verifiable AI safety governance.

🧠 Core Features

• 4D Hypercube Spatial Mapping: Encodes spatial relationships, containment, and state transitions into auditable geometric coordinates.
• Deterministic Alignment Anchors: Designed to supervise, audit, and clamp base model activations to prevent unsafe drift or hallucination.
• Lightweight & Efficient: Fine-tuned via 4-bit quantization, optimized for rapid edge deployment and multi-agent auditing experiments.

🚀 Quick Start: How to Load the Model

You can load your fine-tuned Tesseract LoRA adapters directly into Python using transformers and peft:

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base_model_id = "Qwen/Qwen2.5-1.5B-Instruct"
adapter_id = "TesseractLens-AI/Tesseract-Qwen2.5-1.5B-Spatial"

# Load base model and tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model_id)
model = AutoModelForCausalLM.from_pretrained(base_model_id, load_in_4bit=True)

# Attach your Tesseract Spatial LoRA adapters
model = PeftModel.from_pretrained(model, adapter_id)

print("✅ Tesseract Spatial Safety Supervisor loaded successfully!")