Sovereign Corporate Fiscal Intelligence Transformer v1

Published by ItsNotAI LABS (Dallas, Texas)

The Sovereign-Corporate-Fiscal-Intelligence-v1 is a production-verified PyTorch Multi-Head Self-Attention Transformer model designed for Enterprise Accounting, Tax & Financial Engineering.

🔬 Mathematical Physics & Explicit Parameter Breakdown

Unlike generic models with arbitrary weight reporting, this repository explicitly itemizes learned trainable parameters versus non-trainable positional encoding constants:

Trainable Learned Parameters (requires_grad=True): 4,762,651
Positional Encoding Constant Buffer Elements (pos_encoder.pe): 1,280,000
Total Model State Tensor Elements: 6,042,651
Checkpoint File Size: 23.08 MB
Trained Optimizer: AdamW (10 Epochs over domain datasets)

Governing Mathematical Formulation

$QRE = QRE_{internal} + 0.65 \times Infra + 0.65 \times Contractors$

🎯 Primary Use Cases & Capabilities

Deep Multi-Head Transformer for automated General Ledger expense classification, 12M cash flow forecasting, IRS Form 6765 R&D Tax Credits, and Form 8974 Payroll offsets.
Domain Application: Enterprise SaaS financial operations, IRC §41 & §174 tax credit calculations, and automated ERP chart of accounts assignment.
Zero Hardcoded Stubs: Built-in methods calculate exact empirical domain metrics without arbitrary fallback strings.

📊 Empirical Verification Metrics

Metric	Measured Value
Validation Loss (MSE)	`0.00171`
Empirical Accuracy / Precision	`1.0`
Inference Latency	`6.366 ms`
State Dict Strict Match	`100% PASS`
Dummy Parameter Count	`0`

💻 Python Usage Example

from agent_helper import SovereignCorporateFiscalIntelligenceAgent

# Initialize agent with exact strict state dict loading
agent = SovereignCorporateFiscalIntelligenceAgent()

# Execute domain inference
results = agent.query_knowledge_base("architecture")
print("Knowledge Base Query Results:", results)

⚖️ License

Downloads last month: 10

Safetensors

Model size

6.04M params

Tensor type

F32