UQO Working Paper No. AI-0001· v1.1· November 2026· 77 pages

The Neural Organ Transplant Experiment

Donor-in-the-Loop Activation Grafts Move a Computed Value, Not a Capability, Between Transformers

AI Systems

A one-row activation graft lifts a recipient that never learned an operator from chance to 0.966, but what it moves is the donor's already-computed value, not a capability

Donor and recipient decoder transformers are trained from scratch on an invented operator language, identical except for one operator the recipient never sees; the site where the donor executes it is located by interchange interventions before any transfer, and six graft families are tested against matched controls and conventional learning. The graft needs the donor at inference and is value injection in a code the two models share up to rotation. On Qwen2.5 and Llama-3.2 models, task-vector transfer is supported in 0 of 7 pairs under the frozen rule.

Simon-Pierre Boucher — contact@spboucher.ai

Paper PDF (revised)Read online Code & data repository

0.071 → 0.966

recipient accuracy on held-out Ω programs, from chance to the donor-in-the-loop graft of one orthogonally mapped post-layer-0 row (sd 0.039, min 0.889 over nine donor–recipient pairs); the control stratum stays at 0.992

0.771–0.893

gain over a matched placebo (an unrelated-skill donor or a recipient of another seed through the same map), in 9 of 9 pairs

0.994 vs 0.091

value readout of Ω(v) on the donor's grafted post-layer-0 row against its pre-MLP row: the value is already computed in the row that is grafted

0.993 vs 0.0003

share of items where an operand-swap graft makes the recipient emit the donor's value for the swapped operand, against the value the rule would give for its own operand; pre-MLP and MLP-ablated-donor grafts stay at chance (0.112, 0.104)

−0.456

pre-declared contrast of the donor-free adapter graft against a supervised adapter of the same size; supervised learning or distillation on the same items is perfect, so grafting is not a cheaper way to learn

0 of 7 pairs

open-weight donor–recipient pairs (Qwen2.5-1.5B/3B/7B, Llama-3.2-3B) where task-vector transfer is supported under the frozen rule; a vector-free instruction prefix removes every supported verdict, and the pre-specified query pools are underpowered (minimum detectable differences 0.522–0.585)

up to 0.215

shift of individual pair estimates when the model matrix is retrained on another machine; the pre-specified decision is reproduced in every pair, and a same-hardware re-run is bitwise identical

Abstract

Can a residual-stream state that one transformer uses to execute an operator be grafted into a second transformer that never learned it? We study the question in a micro-world where it is identifiable: decoder transformers trained from scratch on an invented operator language, a recipient that never saw one operator Ω, interchange interventions that locate where the donor executes it, six graft families tested against sham, random, shuffled, unrelated-donor and other-seed controls and against conventional learning, and a decision rule pre-specified in-session. A donor-in-the-loop graft of one mapped post-layer-0 row, through an orthogonal Procrustes map fitted on task-free rows, raises the recipient from chance to 0.966 on held-out programs (control stratum 0.992; placebo-adjusted gain 0.771–0.893 in 9 of 9 pairs). A mechanism battery shows what moves: the donor’s already-computed value, read out in a code the two models share up to rotation (value readout on the grafted row 0.994; an operand-swap graft emits the donor’s value for the swapped operand in 0.993 of items; pre-MLP and MLP-ablated-donor grafts stay at chance). The graft is value injection, not transfer of a computation the recipient executes, and the pre-declared necessity criterion is not Ω-selective. On Qwen2.5-1.5B/3B/7B and Llama-3.2-3B, task-vector transfer is supported in 0 of 7 pairs under the frozen rule, and a vector-free instruction prefix removes every supported verdict (2 one-task partial verdicts remain).

JEL: C45, C63, C88, O33

Key Results

An operational vocabulary for transplantable subsystems

Explicit estimands, a control battery matched on the alignment-map family, a persistence taxonomy, and an elementary proposition on regression-fitted maps whose hypotheses are tested rather than assumed (they fail on these rows).

A micro-world where the question is identifiable

A capability gap manufactured by construction, causal localization before any transfer, exact hook semantics, six pre-specified hypotheses reported with every outcome, a dated deviations table and an append-only registry of every run.

Evidence on the content of the grafted row

Value probes, an operand-swap graft, pre-MLP and MLP-ablated-donor grafts, value-subspace projection and donor-side ablation separate two readings of a successful graft and decide for value injection: the recipient executes nothing it did not already execute, and the subsystem is sufficient but not Ω-selective.

A map of the failure boundaries

Regression-fitted maps fail by relabelling the donor's value onto the recipient's own state; a saved affine adapter scores zero on unseen operands; weight grafts move part of the operator at a cost on the control stratum; steering vectors move nothing; open-weight task-vector transfer does not pass the controls.

A withdrawn control, reported as such

Donor rows from any later layer, grafted through the same map family, work as well as the located row (1.000/1.000 in nine pairs), so the earlier claim that transfer requires the right layer is withdrawn.

Peer review

This paper went through a multi-referee review: 7 referee reports and 116 review items. The version presented on this page is the revised manuscript. Scores are on a 0–100 scale and are a diagnostic of the manuscript, not an acceptance probability.

Score after revision

73.6/100

62.1 before review · +11.5

Review dated November 9, 2026

Review decision: claims not supported as written → reframed and revised
Scientific integrity: PASS
Reproducibility: R1reproduced from documented analysis data
Maturity: Bone revision cycle from submission
Referee reports: 7
Review items: 116

Scores by dimension

Before reviewAfter revision

Originality & contribution65.4 → 67.0
Scientific rigor57.1 → 74.0
Identification & methodology56.9 → 72.0
Data & reproducibility69.0 → 81.0
Robustness & validation64.1 → 78.0
Writing69.3 → 70.0
Figures & tables56.7 → 73.0
Literature positioning62.9 → 78.0

Revised manuscript77 pages17 figures38 tables2 equations63 references35,794 words

What the review changed

Peer review found that the original thesis of capability transfer was not supported as written. The revision adds a mechanism battery showing that the graft injects a value the donor has already computed, re-runs the controls through the primary map, relabels the protocol as pre-specified in-session, and reports every hypothesis outcome, including the open-weight null.

Thesis: capability transfer → value injection in a code shared up to rotation (value readout 0.994; operand-swap graft emits the donor's value in 0.993 of items)
Layer specificity: transfer requires the right layer → withdrawn, since rows from later layers transfer 1.000/1.000 in nine pairs through the primary map
Protocol status: preregistered → pre-specified in-session, with a dated deviations table
Regression-fitted maps: failure by annihilation of directions → failure by relabelling of the donor's value onto the recipient's own state
Conventional-learning comparison: the pre-declared adapter comparator is now reported, giving a negative contrast of −0.456
Open-weight tier: transfer supported in 0 of 7 pairs under the frozen rule; a vector-free instruction prefix removes every supported verdict

Limitations stated in the paper

The micro-world establishes that a computed value can be grafted through a shared code; it does not establish that a computation the recipient would have to execute, a multi-step reasoning state or an open-ended skill can. The toy models are tiny and share their whole training distribution except one operator.
The graft that works depends on the donor at inference; the donor-free variants that would matter for deployment are the weaker ones.
The protocol was fixed in-session on the same trained models used for exploration, with no external timestamp; the design protects against item-level overfitting only.
The open-weight query pools are underpowered for the frozen rule, four models and four tasks do not support claims about families or scales, and results differ between chips.

Read online (interactive reader)Revised PDF Revised branch on spbgit (review-20261103)

Data

Synthetic micro-world

An invented operator language with eight operators, one of which (Ω) the recipient never sees; 17 trained decoder transformers (4 layers, 0.80 M parameters) covering three donor seeds, three recipient seeds, unrelated-skill donors and width, depth and stride variants; evaluation on 1,000 held-out programs with a target stratum, a control stratum and held-out compositions. All data are synthetic.

Open-weight models

Qwen2.5-1.5B, Qwen2.5-3B, Qwen2.5-7B and Llama-3.2-3B instruction-tuned checkpoints, with the capability defined as performing a verifiable task without demonstrations; pre-specified query pools of n = 22–40 and enlarged pools of n = 133–150; seven donor–recipient pairs.

Methodology

Localization before transfer

Interchange interventions on clean and corrupted program pairs, zero-ablation and per-head patching locate the donor's execution of Ω at the post-layer-0 residual row at the read positions.

Alignment maps fitted on task-free rows

The primary map is an orthogonal Procrustes map fitted on 3,000 task-free items; ridge and whitened regression maps on task rows are compared, and a random orthogonal map serves as the sham that defines the estimand.

Six graft families and matched controls

Activation, subspace, steering, adapter, weight and persistent-adapter grafts, against sham-map, norm-matched random, shuffled-item, unrelated-skill donor and other-seed recipient controls routed through the primary map, and against a supervised adapter and full-model distillation on the same items.

Mechanism battery

Value readouts fitted on rows without Ω, an operand-swap graft from a twin program, grafts of the pre-MLP row and of an MLP-ablated donor, value-subspace projection and donor-side ablation of the Ω-contrast directions.

Pre-specified decision rule and inference

A rule fixed in-session before the confirmatory run, paired contrasts against the sham map with a cluster bootstrap over programs, multiplicity adjustment across families, and repetition on a retrained model matrix and on 25 pairs with two new seeds per role (mean 0.879).

Open-weight protocol

Few-shot minus zero-shot last-token task vectors mapped and added at the recipient's best layer, tested against random, unrelated-task and sham-map vectors, a shuffled-query state and a vector-free instruction prefix, under a frozen rule requiring a lower bound above 0.10 on at least two tasks.

Reproducibility

Every run has a manifest with the git commit, per-file code hashes, configuration and split hashes, seeds, library versions and hardware; the registry is append-only and lists every planned, failed, completed and invalidated run.

Figures, tables and inline numbers are regenerated from the committed results by one script; a fresh clone with the archive reproduces every table and number byte-for-byte on another machine.

The trained models, per-item outputs and open-weight caches are archived outside the repository with checksums; open-weight checkpoints are used under their licences and no derived weights are released.

Training determinism across devices is not claimed: a retrained model matrix reproduces the decision in every pair but moves individual estimates by up to 0.215.

Open-weight inference is deterministic per chip but differs between two chips on roughly one cell in six, enough to move one verdict between partial and supported.

Cite this paper

Boucher, S.-P. (2026). The Neural Organ Transplant Experiment. UQO Working Paper No. AI-0001, Université du Québec en Outaouais.

@techreport{boucher2026wpai1,
  author      = {Simon-Pierre Boucher},
  title       = {The Neural Organ Transplant Experiment},
  institution = {Université du Québec en Outaouais, Département des sciences administratives},
  number      = {AI-0001},
  type        = {UQO Working Paper},
  year        = {2026},
  month       = {nov},
  url         = {https://www.spboucher.ai/research/wpai1},
}