MNB Benchmark Observatory Ledger
Public experiment records from the Minimal Navigable Benchmark Observatory.
Overview
The MNB Benchmark Ledger contains 200,000 verified experiment records documenting the evolution of memory-based systems under controlled conditions. Each entry includes cryptographic proof of execution.
Dataset Statistics
| Metric | Value |
|---|---|
| Total Experiments | 200,000 |
| Score Range | 0.3000 - 0.9900 |
| Population Range | 100 - 9,999 |
| Avg Population | 5,053 |
| Verification | 100% |
Data Schema
| Field | Type | Description |
|---|---|---|
experiment_id |
string | Unique experiment ID (mnb_exp_XXXXXX) |
timestamp |
ISO 8601 | Experiment timestamp |
population_size |
integer | Number of MNBs in simulation |
global_coherence_psi |
float | System-wide coherence metric (0-1) |
entropy_h |
float | Information entropy |
mean_value_density_rho |
float | Average value density |
benchmark_score |
float | Composite performance score |
verified |
boolean | Cryptographic verification status |
cryptographic_proof |
JSON | SHA3-256 Merkle proof |
Cryptographic Verification
All entries include SHA3-256 Merkle proofs for reproducibility verification. Each experiment generates a deterministic proof chain enabling independent verification.
from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="MatverseHub/mnb-benchmark-ledger", filename="mnb_ledger_200k.csv")
Usage
import pandas as pd
# Load dataset
df = pd.read_csv("mnb_ledger_200k.csv")
# Filter high-performing experiments
high_score = df[df["benchmark_score"] > 0.8]
# Analyze coherence trends
coherence_trend = df.groupby("population_size")["global_coherence_psi"].mean()
Research Applications
- Memory system performance benchmarking
- Coherence evolution analysis
- Population dynamics modeling
- Cryptographic proof validation
Citation
@dataset{matverse_mnb_benchmark_ledger,
title={MNB Benchmark Observatory Ledger},
author={MatVerse Research Program},
year={2026},
publisher={MatverseHub}
}
Built with 🔐 by MatVerse/Hub