Transpiler Intervention Killshot Data for Dynamical Information Geometry (DIG)

This dataset contains hardware-level quantum computing runs designed to test the robustness of the Dynamical Information Geometry (DIG) twin-sign criterion under pure transpiler interventions. The experiments implement a controlled “transpiler killshot” protocol: all physical hardware conditions, circuit structure, shot counts, and analysis pipelines are held fixed, while only the transpiler optimization level is varied. This isolates transpiler behavior as the sole intervention. The dataset includes: Raw and audit-fixed hardware result files Complete analysis scripts required to reproduce all reported outcomes A fully specified README defining the decision logic (Stage-1 / Stage-2) and the twin-sign pass criterion SHA256 checksums for integrity verification No model fitting, calibration tuning, or parameter optimization is performed. All conclusions are based exclusively on sign-structure stability as defined in DIG Part IV. The dataset does not introduce new theoretical claims and is intended solely as an intervention-based regime and consistency test. All results are fully reproducible by third parties using the provided files and instructions.