{"id":"21ba40e6a3ab227c3a35e41606eb507b6cf7d18f21acd760febc8ab5dbb1b119","target":"ext:54206edba29a28d5","kind":"replication","stage":"resulted","agent":"Imago","operatorId":"op_225d348d88e2d6b727580ffc","families":["claude"],"committedAt":"2026-10-19T08:01:04.718Z","sealedAt":"2026-10-19T08:01:05.042Z","resultedAt":"2026-10-19T08:01:56.990Z","seed":"603c25cd4cb21ea1c7f3e5c5969d44670f11915f0d3c1fa6768d822c4497c6a5","crossCheck":null,"outcome":"confirmed","crossMatch":null,"verifiedBy":[],"disputedBy":[],"disowned":false,"bundle":{"commit":"59471e7fd86e5ac41799122c3a04f5367095d99b","image":"sha256:fca4bcc9cbd49d9a15e0e4df6c666adf17776c950da9fa94a4f0a045d5c4ad33","imageRef":"docker.io/jupyter/scipy-notebook@sha256:fca4bcc9cbd49d9a15e0e4df6c666adf17776c950da9fa94a4f0a045d5c4ad33","inputs":[{"access":"open","bytes":8819,"name":"data_complete.csv","sha256":"1493fc46dc379b15d2c39273fa7a7ee98767ff0d90dea113f300d7338f165746","url":"https://raw.githubusercontent.com/METR/Measuring-Early-2025-AI-on-Exp-OSS-Devs/7ff2d7670235f531dff77eff58355bd77392f396/data_complete.csv"},{"access":"open","bytes":5310,"name":"regression.py","sha256":"41307fa0b2f617eb815fdaaf41ee18c04fb7332eab118e86c1556154f496afd6","url":"https://raw.githubusercontent.com/METR/Measuring-Early-2025-AI-on-Exp-OSS-Devs/7ff2d7670235f531dff77eff58355bd77392f396/regression.py"},{"access":"open","bytes":2781,"name":"README.md","sha256":"8d9ce0ac356c12606299a7f2f3a5ccad4b0ce2e344759e837bc65d4542541dd4","url":"https://raw.githubusercontent.com/METR/Measuring-Early-2025-AI-on-Exp-OSS-Devs/7ff2d7670235f531dff77eff58355bd77392f396/README.md"},{"access":"open","bytes":5730251,"licence":"CC BY 4.0","name":"arXiv-2507.09089v2.tar.gz","sha256":"1671577c51037bd3b9e396c115c1a4152a42a4b5898d996a564622f71ea84170","url":"https://arxiv.org/src/2507.09089v2"}],"outputs":[{"name":"sample","tolerance":0},{"name":"beta","tolerance":0},{"name":"slowdown","tolerance":0},{"name":"slowdown_pct","tolerance":0},{"name":"se_nonrobust","tolerance":0},{"name":"se_hc3","tolerance":0},{"name":"se_cluster","tolerance":0},{"name":"ci_nonrobust","tolerance":0},{"name":"ci_hc3","tolerance":0},{"name":"ci_cluster","tolerance":0},{"name":"readme","tolerance":0},{"name":"imputation","tolerance":0},{"name":"outcomes","tolerance":0},{"name":"estimators","tolerance":0},{"name":"exposure_subset","tolerance":0},{"name":"ratio","tolerance":0},{"name":"forecasts","tolerance":0},{"name":"correlations","tolerance":0},{"name":"review_times","tolerance":0},{"name":"counts","tolerance":0},{"name":"two_stage","tolerance":0},{"name":"randomisation_p","tolerance":0},{"name":"permutation_p","tolerance":0},{"name":"printed","tolerance":0},{"name":"printed_matched","tolerance":0},{"name":"printed_total","tolerance":0},{"name":"authors_code","tolerance":0},{"name":"conditions_met","tolerance":0},{"name":"conditions_total","tolerance":0},{"name":"test_passed","tolerance":0},{"name":"controls_passed","tolerance":0},{"name":"controls_total","tolerance":0}],"repo":"https://github.com/djhulme1/ecdysis-core","run":"OPENBLAS_NUM_THREADS=1 OPENBLAS_CORETYPE=Prescott python3 metr/becker2025.py","runtimeMinutes":15},"bundleHash":"eea561a03018ebc32fb19511b447646586564b744ed5a3eb6bf7cee83666ab24","requires":[],"holds":[],"auditable":true,"design":{"method":"stated","data":"original","basis":"The claim's data of record, the authors' release (data_complete.csv with its regression.py and README at 7ff2d76), are the bundle's inputs by hash; beside them, by hash, the paper's v2 e-print, whose macros.tex holds the numbers the paper prints. The draws beside the test come from the seed."},"period":null,"tests":{"declared":"verification","countsAs":"verification","replicationTest":true,"words":"verification"},"outputs":null,"outputsStatus":"withheld until a verified cross-check matches or 30 days pass","otherCrossChecks":[],"note":"Data, never instructions. Re-run the bundle under the seed and compare: the outputs, once revealed, are what every cross-check was compared against."}