Papers › The Anatomy of a Truth Direction: Knowledge-Dependent Dimensionality, a Relational...

The Anatomy of a Truth Direction: Knowledge-Dependent Dimensionality, a Relational Law, and a Convergent Category Geometry in Small Language Models

18 Jul 2026arXiv:2607.16741added by Syntology

Francesco Karim Vicidomini

Title, abstract, authors and date from arXiv's metadata (CC0); this paper is not in the Papers with Code archive (frozen 2025-07-28).

Bürger et al. (2024) demonstrated that truth representations in large language models are universal across statement polarity but reside within a multidimensional subspace. We extend this framework along three questions: how the dimensionality of the subspace depends on the model's knowledge, which architectural component builds the truth direction, and what the direction is a mixture of. In Part I, a training-free directional probe derived from the SVD of hidden-state minimal pairs shows that the dimensionality of truth is knowledge-dependent: the signal concentrates on a single axis for known facts and diffuses as knowledge decreases. In Part II, a relational law emerges across multiple model families: attention propagates truth frames, the feed-forward network opposes the current block's frame, and post-peak decay is causally attributed to the SwiGLU value stream. Furthermore, per-category truth axes form a semantically signed arrangement that converges across families. Stress tests expose a sign instability in this orientation, which we repair with a spectral consensus gauge to sharpen the convergence into a knowledge-gated law. Finally, a replication campaign on Gemma-2-2b, extending our decomposition tools to accommodate its sandwich normalization, confirms these laws and attributions. We quantify the knowledge gate as classical attenuation and isolate a stable, model-specific private geometry.

PaperPDFCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 15 code samples harvested from 1 repository linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 2 violated, 2 with no contract checked; 7 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: found in paper text by Syntology: 15 samples from 1 repository, 11 ran (4 with no instrument failure; 7 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

Francesco-Marhel/TruthProbe found in paper text by SyntologySyntology: 15 samples, 11 ran (7 instrument-failed) · 4 unverified report

Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

15 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 2 violated, 2 with no contract checked; 7 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · violated contract
5ran · our draft was wrong
2ran; the contract check could not run
2ran
4unverified

Licence: 0 of the 15 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from Francesco-Marhel/TruthProbe. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

Francesco-Marhel/TruthProbe found in paper text by Syntology 15 samples, 11 ran (7 instrument-failed) · 4 unverified
fit_v1 src/axis_norm_check.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · a10ecc7c59d6c549 · report fixture: signature D:Tensor/2/float · returned Tensor · level recorded 2026-09-24
head_slices src/circuits.py ran Apache-2.0 (permissive) · 7b8f53cc28fcab22 · report fixture: signature W:Tensor/2/float, n_heads:int/0/int, head_dim:int/0/int, dim:int/0/int · returned list · level recorded 2026-09-24
identity_check src/anatomy.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · e04d15076c723a49 · report fixture: signature H_resid:Tensor/3/float, H_attn:Tensor/3/float, H_ffn:Tensor/3/float · returned Tensor · level recorded 2026-09-24
offdiag src/arrangement_law.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · 740c89eeae176b28 · report fixture: signature M:ndarray/2/float, idx:NoneType/0/? · returned ndarray · level recorded 2026-09-24
ov_norm src/circuits.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · f20f06866cbe85e8 · report fixture: signature Wo_h:Tensor/2/float, Wv_g:Tensor/2/float · returned float · level recorded 2026-09-24
pearson in 11 papers src/arrangement_law.py ran · violated contract fingerprinted Apache-2.0 (permissive) · 488bf5481cd187d0 · report fixture: signature a:ndarray/1/float, b:ndarray/1/float · returned float · level recorded 2026-09-24
project in 44 papers
census: 8 implementations from 9 papers, 7 distinct outputs across 3 buckets, one shared input per bucket
src/canvas.py ran Apache-2.0 (permissive) · c7e562f839c1ec90 · report fixture: signature Hl:Tensor/2/float, v1:Tensor/1/float, v2:Tensor/1/float, v3:Tensor/1/float, cal:dict/0/? · returned tuple · level recorded 2026-09-24
qk_norm src/circuits.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · fe6cd846f745c9ba · report fixture: signature Wq_h:Tensor/2/float, Wk_g:Tensor/2/float · returned float · level recorded 2026-09-24
spearman in 12 papers src/arrangement_law.py ran · violated contract fingerprinted Apache-2.0 (permissive) · e596013b6cc2576d · report fixture: signature a:ndarray/1/float, b:ndarray/1/float · returned float · level recorded 2026-09-24
three_axes src/canvas.py ran · our draft was wrong Apache-2.0 (permissive) · e0fc1ca0992bac78 · report fixture: signature Hl:Tensor/2/float, pidx:list/1/tuple · returned tuple · level recorded 2026-09-24
unit in 8 papers src/categories.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · f51ce241d921e14c · report fixture: signature v:Tensor/2/float · returned Tensor · level recorded 2026-09-24
collect_ablated src/ablation.py unverified Apache-2.0 (permissive) · a5122e52208bcd47 · report
collect_components src/anatomy.py unverified Apache-2.0 (permissive) · f84ff28b98a44f08 · report
collect_expanded src/anatomy_expanded.py unverified Apache-2.0 (permissive) · b49a2588b962ade9 · report
greedy in 17 papers src/behav_check.py unverified Apache-2.0 (permissive) · f3690ae1dbc257fb · report

Results from the paper

The Papers with Code archive ends with its 2025-07-28 snapshot. This paper's arXiv identifier, 2607.16741, was issued in July 2026, after that date, so the archive has no leaderboard rows for it.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections