| --- |
| license: cc-by-4.0 |
| tags: |
| - interpretability |
| - jacobian-lens |
| - pythia |
| - reproducibility |
| library_name: pytorch |
| --- |
| |
| # Corpus and context dependence of the Jacobian lens — artifact mirror |
|
|
| Fitted operators, experimental records and the corpus pool for the accompanying paper. |
|
|
| **Licence.** Everything here is **CC BY 4.0**, without qualification — see `LICENSE`. That is |
| true by construction, not by assertion: this repository contains only artifacts this work |
| produced. **No third-party corpus text is distributed here.** The corpus pool is rebuilt from its |
| public sources by `repro/06_build_corpora.sh` and verified against the manifests that ship here; |
| 8 of 11 corpora rebuilt bit-identically when measured. `ATTRIBUTION.md` has the detail. |
|
|
| ## Layout |
|
|
| ``` |
| results/lenses/e28/ 166 operators + sidecars corpus x N ladder |
| results/lenses/plad/ 53 operators + sidecars per-layer activation dumps |
| results/lenses/ladder/ 15 operators + sidecars canonical model ladder (70m ... 2.8b) |
| results/lenses/misc/ 8 operators smokes and one-offs |
| results/<17 experiment dirs> per-experiment operators and records |
| results/*.json 190 experimental records |
| corpora/ 3 provenance manifests (NO corpus text — rebuild it, see ATTRIBUTION.md) |
| ``` |
|
|
| Every `.pt` is a corpus-averaged Jacobian `J_l = E_x[dh_final/dh_l]` fitted with the reference |
| implementation at pin `581d398`, fp32, TF32 off, uncentered. |
|
|
| ## Reproducing |
|
|
| Every statistic in the paper recomputes from `results/**.json` alone — no operators, no GPU: |
|
|
| ```bash |
| git clone <code repository> |
| uv sync && bash repro/01_setup_local.sh |
| bash repro/04_fetch_results.sh # pulls results/**.json from here, ~7 MB |
| bash repro/RUN_ALL.sh --tier T0 # 12 modules |
| ``` |
|
|
| `--tier T1` re-scores the stored operators on CPU and needs the `.pt` from this repo |
| (`bash repro/03_fetch_artifacts.sh --all`). `--tier T2` refits on GPU. |
|
|
| ## Provenance |
|
|
| Each results file carries a provenance block: the producing script and its SHA-256, the commit, |
| the argv, and the SHA-256 of every declared input. `payload_sha256` covers the record **excluding** |
| the provenance block, so it is stable across reruns that change only environment metadata — rerun |
| and compare it, and a match means no number moved. |
|
|
| No hostname is recorded. Input hashes are the values at write time; where a record was later |
| redacted, `payload_sha256_before_redaction` retains the original so both remain verifiable. |
|
|
| No corpus plaintext is distributed — see `ATTRIBUTION.md` for the rebuild. |
|
|
