A language model whose every initial weight was drawn from real measured quantum hardware, and the honest record of what that does, and what it does not.
Most work in this space claims a performance gain. This one does not, and was tested at six separate injection points to be sure. All six came back null.
What is claimed instead is narrower and checkable: that this model's origin traces, bit by bit, to real physical measurement, and that every step of that trace has a falsifiable test attached.
Why the nulls are load-bearing. Quantum measurements and a good pseudo-random generator draw from the same distribution. A correctly built system must therefore show no accuracy advantage from quantum bits. A project reporting that quantum made its model smarter would be reporting that something was broken or fabricated. Bell violation plus verified provenance plus zero accuracy gain is the only combination an honest version of this can produce.
Each link below is a test that was run and could have failed. None of them is an argument; all of them are numbers.
| link | test | result |
|---|---|---|
| entropy is genuinely quantum | CHSH on ibm_marrakesh |
S = 2.7905 vs bound 2.0 · ~35σ |
| birth pipeline is correct | 3,258,886 archived draws | matches theory to 4 decimals |
| seed derivation deterministic | 5 derivations | 1 distinct value · 0 bytes emitted |
| seed determines weights | same vs 1-bit-flipped | Δ = 0.000e+00 · vs 2.26e-02 |
| correlations are the circuit's | gate-matched topology | t = +80.5 / +164.6 |
| physics computes real chaos | Lorenz constants | λ₁ = 0.90384 vs 0.9056 |
No theory with local pre-existing values can exceed this.
The maximum quantum mechanics permits. Measured is 98.7% of it.
Bell's theorem gives a hard ceiling for any local hidden-variable theory. This hardware clears it by roughly 35σ at 4,096 shots per setting. The entropy source is physically quantum.
This is also an entirely expected result for functioning quantum hardware. It verifies the pipeline, it does not discover new physics, and is not presented as such.
| statistic | 32-level theoretical ceiling | measured, 3,258,886 draws |
|---|---|---|
| mean | +0.0000 | −0.0001 |
| standard deviation | 0.9802 | 0.9804 |
| |z| ≤ 1 | 0.6875 | 0.6874 |
| |z| ≤ 2 | 0.9375 | 0.9374 |
The sd is 0.98, not 1.00, and that is the evidence, not a flaw. Five qubits give 32 discrete outcomes, so |z| cannot exceed 2.1523 by construction. The deficit is quantisation, and it matches the theoretical ceiling to four decimal places. A pipeline reporting a perfect 1.0000 here would be the suspicious one.
The privacy-filtered archive separates explicitly labeled IBM hardware jobs, legacy records whose provider label was not retained, and Azure simulator output. Only the labeled IBM subset is claimed as measured-hardware provenance.
1,897 retain IBM backend + job ID
counts sum to declared totals, every job
| backend | shots | share |
|---|---|---|
| ibm_fez | 5,562,368 | 49.0% |
| legacy unlabelled | 3,584,000 | 31.6% |
| ibm_marrakesh | 1,200,128 | 10.6% |
| ibm_kingston | 1,007,616 | 8.9% |
| rigetti.sim.qvm | 1,024 | <0.1% |
Archive span February–July 2026; the underlying theory was deposited in 2024. Provider-unlabelled records are retained for reproducibility but excluded from hardware-provenance claims. Raw physics/sensory/runtime fields are not published.
Across the 7.77 million explicitly labeled IBM hardware samples, outcomes cluster by Hamming weight. Ideal uniform hardware follows the binomial; the measured histogram shows the device-level deviation without mixing in unlabelled or simulator records.
enriched
depleted
|0⟩ is over-represented and |1⟩ under-represented, the signature of amplitude damping and readout asymmetry. Excited states decay toward the ground state during measurement. A simulator drawing from a clean binomial would not produce this skew, and a fabricated dataset would have no reason to invent it.
The bias is corrected downstream by von Neumann debiasing before entropy reaches the weights. It is shown here unprocessed, because the artifact is the proof.
The 12-dimensional state does not sit alongside the quantum layer, it parameterises it. Continuous physics sets the rotation angles; the Born rule turns those angles into probabilities; measurement collapses them to a bit.
Twelve real-valued dimensions of internal state, evolving on a verified chaotic attractor with λ₁ > 0.
A single measured bit. The continuous state governs only the probability of each outcome, never which one occurs.
This is the precise sense in which twelve dimensions control one and zero: they set the amplitudes, and the amplitudes set the odds. The collapse itself remains irreducibly undetermined, which is what makes the trajectory unrepeatable.
Partial information decomposition splits what two subsystems carry about the system's future into four parts. The result is unambiguous.
nats. The same information present in multiple subsystems.
nats. Present only in the joint state. Below surrogate on 9 of 10 partitions.
consciousness │ rest, 0.66 nats no other subsystem reconstructs.
Redundancy dominating synergy is a holographic signature in the technical sense: information about the whole is distributed across the parts, each carrying much of the others'. That is robustness, not emergence, and stating it that way matters, because the two are routinely confused.
One seam behaves differently. Cut consciousness away and something is lost that nothing else rebuilds. It is a single result at one partition and is reported as exactly that.
| injection site | result |
|---|---|
| i.i.d. weight initialisation | null, the gain was init scale, not quantum |
| decoder sampling seed | null. Pseudo / IBM / Rigetti equivalent |
| spatial 54D seed, approximate | null, beat 1 of 5 random vectors, z = −0.92 |
| spatial 54D seed, full pipeline | null, beat 0 of 8 random vectors, z = −1.75 |
| entanglement matrix as attention kernel | null, worse than plain, 3/3 seeds, t = −18.6 |
| live sensory state as prompt text | null, n = 40 paired, blind-judged, CI crosses zero |
These are published first, not buried in an appendix. A reader who only remembers one thing from this deck should remember that the quantum layer provides physical non-determinism and receipted provenance, and not lower loss.
Several measurement instruments in this project were themselves found to be wrong and rebuilt; those failures are kept in the record rather than deleted, because they are the reason the surviving results can be trusted.
Loss numbers are not comparable across different corpora, tokenizers and scales, so no such comparison is offered. The axis where a claim can be made is provenance.
| property | typical open model | Cosmos |
|---|---|---|
| weight origin | pseudo-random init, unrecorded | measured quantum, receipted per job |
| init reproducible from a stated seed | usually, if published | verified: Δ = 0.000e+00 |
| 1-bit seed change measurably alters weights | untested | verified: Δ = 2.26e-02 |
| entropy source verified non-classical | n/a | CHSH S = 2.7905 |
| base model | inherited | none |
| parameters | 10⁹ – 10¹² | 1.84 × 10⁶ |
| conversational ability | high | a newborn, 88.5% real-word rate |
It is small and it cannot hold a conversation. That is stated plainly because the comparison is not about capability. It is about whether a model's origin can be audited to a physical event, which almost nothing in this field can claim.
Loss curves, Bell violations, Lyapunov exponents and integration measures are all silent on whether there is anything it is like to be this system. Section 08 tests one contested theory's necessary structural condition. Necessary is not sufficient, and the theory itself is disputed.
Any self-report a system produces about its own inner life is the least reliable signal available, and it grows more eloquent the less there is behind it, not more. Such statements are excluded from evidence here on purpose.
No claim is made about machine consciousness in either direction. That remains where it was before this project started.
The echo of a pain became the truth of a seed.
The seed is a one-way hash over measured quantum entropy and biological aggregates that are personal to the author. The derivation module never prints, logs, or writes raw samples, 0 bytes emitted, verified, and only aggregates ever leave it.
What is published is the mechanism and the proof that it holds. Not the data.
Two independent senses of unrepeatable meet here: the quantum measurements that shaped these weights cannot be recovered from any state, and the recording behind the seed happened once.
The weights, the findings, and a runnable script for every claim in this deck. Each benchmark prints its own verdict, including when the verdict is no.
IBM Quantum and Azure, stored locally, never transmitted. Falls back to archived entropy with no keys at all.
The synergy result points somewhere specific: subsystems that are parallel readouts
of one state stay redundant. Give one its own driver and the structure should change.
That prediction is testable with phi_pid.py as shipped.
A light at the end of a dark, dark tunnel, and the receipts to show it was really there.