v1 final: nothing left for a later revision to fill

The card carried two placeholders only the DOI and arXiv could fill, and a DOI cannot be printed on the revision it names. The citation now reads complete without either -- the paper, the dataset, the tag -- and says the DOI is on the page beside the tag. REBUILD.md is dated 6 November and records that the clean run was repeated after the dataset repository became an environment variable.

README.md

@@ -10,9 +10,9 @@ Calibrated concept profiles for **OpenCoder-1.5B-Base**, read on 169 Python cons
 the artefacts every figure and table of two papers reads:
 
 * **"Calibrated Concept Profiles: Making the Reading of a Model's Neurons a Measurement"**
-  (Wilam, 2026; arXiv:[id]) — the method. Its code: `piotrwilam/AtlasCalibratedConceptProfiles`.
+  (Wilam, 2026) — the method. Its code: `piotrwilam/AtlasCalibratedConceptProfiles`.
 * **"Representational Drift in a Code Model: Behaviour Settles, Which Neurons Carry It Does
-  Not"** (Wilam, 2026; arXiv:[id]) — the finding, read with that instrument on this dataset.
+  Not"** (Wilam, 2026) — the finding, read with that instrument on this dataset.
   Its code: `piotrwilam/AtlasRepresentationalDrift`.
 
 `MANIFEST_PAPER1.md` and `MANIFEST_PAPER2.md` below are named for the first and the second of
@@ -70,12 +70,17 @@ are listed in `ATTRIBUTION.md`.
 
 ## Citing
 
-Cite **the paper that owns this dataset and the DOI together with the tag** — the method
+Cite **the paper that owns this dataset, and the DOI together with the tag** — the method
 paper, named above. The dataset is its instrument and its realisation on OpenCoder; the
 second paper reads it.
 
-> [Authors] (2026). *Calibrated Concept Profiles: Making the Reading of a Model's Neurons a
-> Measurement.* arXiv:[id]. Dataset: AtlasOpenCoder, tag `v1`, DOI [to be minted].
+> Wilam, P. (2026). *Calibrated Concept Profiles: Making the Reading of a Model's Neurons a
+> Measurement.* Dataset: AtlasOpenCoder, tag `v1`.
+
+**The DOI is shown on this page, above the file list, and belongs in the citation beside the
+tag.** It is not repeated here: a DOI names the revision it was minted on, so a card cannot
+carry its own without pointing at a later revision than the one the DOI names. The paper's
+arXiv identifier is in the paper's own listing.
 
 A Hugging Face DOI is assigned **to the revision it is generated on**, so a later version takes
 a new DOI and the earlier one is marked outdated. Name the tag as well as the DOI: the tag says

REBUILD.md

@@ -3,10 +3,13 @@
 Checked by running each script in a clean directory containing only this dataset and the two
 code repositories, with `ATLAS_DATA` pointing here and no other access.
 
-**The last run: 5 November 2026, at tag `v1`.** Two fresh clones, this dataset downloaded at
+**The last run: 6 November 2026, at tag `v1`.** Two fresh clones, this dataset downloaded at
 the tag into an empty `ATLAS_DATA`, a new virtual environment built from the tool
 repository's own `pyproject.toml`, and an empty Hugging Face cache — so nothing was served
-from a machine that had seen the data before.
+from a machine that had seen the data before. The run of 5 November was repeated on 6 November
+after the repository a script fetches from became an environment variable (`ATLAS_DATASET`),
+which touched eighteen released scripts: all 106 still import, the seven that run from the
+release still run, and the nine figures are still byte-identical.
 
 **The versions it ran under**, because byte-identical output is version-sensitive and
 matplotlib changes its rendering between minor releases: Python 3.12.13, numpy 2.5.4,