| { |
| "passed": true, |
| "environment": "ising-decoding", |
| "interpreter": "Python 3.12.13 (ising-decoding validation environment)", |
| "weights": { |
| "passed": true, |
| "v2_models_strict_loaded": 9, |
| "tensor_prediction_checks": 12, |
| "native_rounds": [ |
| 10, |
| 30, |
| 50 |
| ], |
| "maximum_absolute_logit_difference": 0, |
| "legacy_v1_strict_load": true, |
| "torch": "2.5.1+cu121" |
| }, |
| "patch": { |
| "upstream_commit": "33acb152e403bc189f2effdb07f1a87b34c745f1", |
| "patch_sha256": "b73f8b4a5ce40066f33dabd3102a438c6e3045c42cb1e5f282abb9daabd79c68", |
| "patch_applied_to_clean_upstream": true, |
| "patched_files_byte_identical": 33 |
| }, |
| "examples": [ |
| { |
| "example": "paired_inference_compare", |
| "passed": true, |
| "mode": "cpu_smoke_64_shots_per_basis", |
| "jobs": 0 |
| }, |
| { |
| "example": "infer_t0_t4", |
| "passed": true, |
| "mode": "dry_run", |
| "jobs": 5 |
| }, |
| { |
| "example": "infer_ood", |
| "passed": true, |
| "mode": "dry_run", |
| "jobs": 396 |
| }, |
| { |
| "example": "infer_willow", |
| "passed": true, |
| "mode": "dry_run", |
| "jobs": 1 |
| } |
| ], |
| "native_willow_smoke": { |
| "passed": true, |
| "patch": "d3_at_q4_5", |
| "basis": "X", |
| "rounds": [ |
| 10, |
| 30, |
| 50 |
| ], |
| "shots_per_case": 16, |
| "device": "cpu" |
| }, |
| "safetensors_regression_tests": { |
| "passed": 14, |
| "reused_unchanged_code_result": true |
| }, |
| "data": { |
| "source_paper_data_sha256": "9f09e53bb0e26178e2ad2f5e9789cab9302bdf3502d5d0e21e5969d66fc9f565", |
| "published_paper_data_sha256": "368ae81662ac3ad6b384857dec406b2a9e050ecd26fdabdf55c5e8edcc1802c2", |
| "filter": "Removed standalone PyMatching rows and comparisons; retained all non-PyMatching values without recalculation.", |
| "json_records_removed": 10422, |
| "csv_filters": { |
| "terminal_metrics.csv": { |
| "rows_kept": 6, |
| "rows_removed": 0 |
| }, |
| "ood_cells.csv": { |
| "rows_kept": 9504, |
| "rows_removed": 792 |
| }, |
| "willow_cases.csv": { |
| "rows_kept": 672, |
| "rows_removed": 84 |
| } |
| }, |
| "headline_metrics_unchanged": true, |
| "ood_conditions": 396, |
| "willow_cases": 84 |
| }, |
| "verification_scope": "Checkpoint tensor equality, exact reference predictions, clean-patch reconstruction, CLI dry runs, and small CPU decoding checks. Full research evaluations were not rerun." |
| } |
|
|
