It was run inside an isolated container with no network access. This is the exact command and the real output it produced — captured process output, not written by a model.
$ python3 hmsm.py
────────────────────────────────────────────────────────────────────────
HMSM — demo mode (80 synthetic data points)
────────────────────────────────────────────────────────────────────────
{
"themes": [
{
"theme_id": 0,
"size": 19,
"centroid": [
0.02631578947368421,
0.02631578947368421,
0.02631578947368421,
0.02631578947368421,
0.02631578947368421,
0.02631578947368421,
0.0,
0.030386856273138203,
0.030386856273138203,
0.030386856273138203,
0.030386856273138203,
0.030386856273138203,
0.030386856273138203,
0.0,
0.030386856273138203,
0.030386856273138203,
0.030386856273138203,
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,
0.02631578947368421,
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,A clean run proves this does what is shown above, in a CPU-only sandbox. It is a small research demo — not a production tool, and nothing here was published anywhere.