Scruff AI
56c71c87b2
- cuda/ — main LBM kernel (khra_gixx_1024_v5.cu) - navigator/ — lattice_observer, golden_weave, bridges, mock daemon - scripts/ — compile, start, launch, setup (paths updated) - docs/ — system manual - archive/ — everything else (old kernels, inquiries, experiments) - README.md — full setup guide: requirements, quick start, use your own LLM - removed training/ entirely (broken LoRA scripts + datasets) - .gitignore: exclude build/ logs/ training/ *.jsonl
29 lines
1.0 KiB
Python
29 lines
1.0 KiB
Python
import re
|
|
|
|
with open("/tmp/khra_daemon.log") as f:
|
|
lines = f.readlines()
|
|
|
|
vals = []
|
|
for l in lines[-100:]:
|
|
m = re.search(r"Cycle (\d+): Coherence=([0-9.]+), Asymmetry=([0-9.]+)", l)
|
|
if m:
|
|
vals.append((int(m.group(1)), float(m.group(2)), float(m.group(3))))
|
|
|
|
if vals:
|
|
coh = [v[1] for v in vals]
|
|
asym = [v[2] for v in vals]
|
|
mean_c = sum(coh) / len(coh)
|
|
mean_a = sum(asym) / len(asym)
|
|
std_c = (sum((c - mean_c) ** 2 for c in coh) / len(coh)) ** 0.5
|
|
std_a = (sum((a - mean_a) ** 2 for a in asym) / len(asym)) ** 0.5
|
|
print(f"LAST {len(vals)} SAMPLES (cycles {vals[0][0]} to {vals[-1][0]})")
|
|
print(f" Coherence: min={min(coh):.4f} max={max(coh):.4f} mean={mean_c:.4f} stdev={std_c:.6f}")
|
|
print(f" Asymmetry: min={min(asym):.4f} max={max(asym):.4f} mean={mean_a:.4f} stdev={std_a:.6f}")
|
|
print()
|
|
print("FIRST 5:")
|
|
for v in vals[:5]:
|
|
print(f" Cycle {v[0]:>6}: C={v[1]:.4f} A={v[2]:.4f}")
|
|
print("LAST 5:")
|
|
for v in vals[-5:]:
|
|
print(f" Cycle {v[0]:>6}: C={v[1]:.4f} A={v[2]:.4f}")
|