# Sources and provenance

Sources checked 2026-09-07. Literature supplies methods and comparison points, not validation of this local experiment. The following descriptions are intentionally brief; the report's numerical findings come from local saved results.

- [J-space paper](https://transformer-circuits.pub/2026/workspace/index.html): defines an averaged Jacobian transport followed by normalization/unembedding, and a pseudoinverse-based swap in a nonorthogonal span. Readouts are suggestive; interventions test effects. Our experiment does not test the paper's broader global-workspace account.
- [Reference implementation](https://github.com/anthropics/jacobian-lens/tree/581d398613e5602a5af361e1c34d3a92ea82ba8e): frozen in `vendor/jacobian-lens`, commit `581d398613e5602a5af361e1c34d3a92ea82ba8e`. Read `hf.py`, `hooks.py`, `lens.py`, `fitting.py`, tests, README and the walkthrough. It records zero-based block outputs, transports row vectors with `h @ J.T`, and applies the original final norm and unembedding. README explicitly says reference code is unmaintained. Apache-2.0.
- [Walkthrough](https://github.com/anthropics/jacobian-lens/blob/581d398613e5602a5af361e1c34d3a92ea82ba8e/walkthrough.ipynb): Qwen3.5-4B, bf16 CUDA, `qwen-n1000`, boot/currency example. This example is used only as a local smoke check, never evaluation evidence.
- [Neuronpedia lens artifacts](https://huggingface.co/neuronpedia/jacobian-lens/tree/16a01f309fcec900fdcec3f4cd5b64f3d00e4d5a): Qwen3.5-4B n1000 file exists, 406,332,644 bytes; model shards total 9,319,828,096 bytes. Download not selected given free disk. Qwen3-1.7B lens is 226,501,315 bytes with 466 fitted prompts (despite the repository revision name). Fit configuration and model history are saved locally.
- [Qwen3-1.7B checkpoint](https://huggingface.co/Qwen/Qwen3-1.7B/tree/70d244cc86ccca08cf5af4e1e306ecf908b1ad5e): full original safetensors and tokenizer, no quantization/conversion. Model is 28 blocks, residual width 2048, tied embedding/unembedding, learned RMSNorm gain. Apache-2.0. Artifact SHA256s and sizes are in `provenance/artifacts.json`.
- [Neuronpedia fitting wrapper](https://github.com/hijohnnylin/neuronpedia/blob/df03cfee33e9023866eba6f828d9e34900932877/utils/neuronpedia-utils/neuronpedia_utils/jlens/fit_lens.py): downloaded at this commit. Delegates model wrapping, fitting and saving to jlens. The fit YAML does not record a model commit or historic jlens commit. Model repository history establishes that the latest checkpoint/tokenizer predates the 2026-06-11 fit; it does not recover the exact historic fitting environment. Treat that missing historic code provenance as a limitation.
- [CHIVE](https://alignment.anthropic.com/2026/chive/), 2026-08-21: counterfactual prompt edits provide measured labels; AO/NLA/SAE readout tools did not improve prediction over transcript access in the tested settings. It did not evaluate this local J-lens intervention experiment. Motivates separating readout visibility from useful counterfactual prediction.
- [Circuit tracing methods](https://transformer-circuits.pub/2025/attribution-graphs/methods.html): cross-layer transcoders create a replacement model for attribution graphs; causal checks remain necessary because the replacement is approximate. We fit no transcoder and recover no attribution graph.
- [Circuit tracing case studies](https://transformer-circuits.pub/2025/attribution-graphs/biology.html): Dallas/Texas/Austin and California/Sacramento illustrate intervention on intermediate features, alongside a Dallas-to-Austin shortcut. This motivates our two-step task; none of those cases enter local evaluation.
- [Introspection experiments](https://www.anthropic.com/research/introspection): controlled concept injection can sometimes be detected/reported under specific conditions. It does not validate ordinary post-answer explanations as records of the preceding computation.
- [Apple PyTorch/Metal](https://developer.apple.com/metal/pytorch/): MPS uses the Mac GPU. The page currently lists PyTorch 2.11.0 in its installation example; the package index installed 2.14.0 here. We record the actually installed version and execute checks instead of treating the example as a version pin.
- [MLX documentation](https://ml-explore.github.io/mlx/build/html/index.html), [lazy evaluation](https://ml-explore.github.io/mlx/build/html/usage/lazy_evaluation.html), [eval](https://ml-explore.github.io/mlx/build/html/python/_autosummary/mlx.core.eval.html), [MLX-LM](https://github.com/ml-explore/mlx-lm): GPU arrays/lazy computation require explicit evaluation and synchronization for timing. Installed MLX-LM Qwen3 source was inspected for post-block residual access and tied unembedding.
- [Core ML prediction](https://apple.github.io/coremltools/docs-guides/source/model-prediction.html) and [typed execution](https://apple.github.io/coremltools/docs-guides/source/typed-execution.html): converted prediction graphs may be partitioned across CPU/GPU/Neural Engine and precision depends on conversion/runtime. Inference support is insufficient evidence of access to the required intervention boundary. We did not convert a model or execute on the Neural Engine.
- [SIX ISO 4217 list](https://www.six-group.com/dam/download/financial-information/data-center/iso-currrency/lists/list-one.xml): downloaded to `provenance/iso4217-list-one.xml` to cross-check currency codes. Cities are deliberately constructed familiar factual examples; no claim of random sampling or independently audited geography dataset.

Editorial references were read from the requested absolute paths and remain unchanged; hashes are in `provenance/editorial-inputs.json`. Generated prose is this investigation's interpretation, not the papers' conclusion about this model.