The card names both papers by title

A reader arriving here could not tell which papers 'the two papers' were, nor which manifest belonged to which. Both are now named in full with their code repositories, and the two manifests are said to be named for the first and the second in that order. The arXiv identifiers are slots, filled on the day.

README.md

@@ -7,7 +7,16 @@ pretty_name: AtlasOpenCoder
 
 Calibrated concept profiles for **OpenCoder-1.5B-Base**, read on 169 Python constructs across
 61 pretraining and annealing checkpoints. This dataset is the battery, the licence tables and
-the artefacts every figure and table of the two papers reads.
+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`.
+* **"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.
+  Its code: `piotrwilam/AtlasRepresentationalDrift`.
+
+`MANIFEST_PAPER1.md` and `MANIFEST_PAPER2.md` below are named for the first and the second of
+those, in that order.
 
 ## What is here
 
@@ -61,8 +70,9 @@ are listed in `ATTRIBUTION.md`.
 
 ## Citing
 
-Cite **the paper that owns this dataset and the DOI together with the tag**. The dataset is the
-instrument of the first paper and its realisation on OpenCoder; the second paper reads it.
+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].