{"schema_version":1,"total":1,"offset":0,"limit":100,"observations":[{"id":"public:55a84368f006dde9d6900b85","benchmark_id":"mystery-game-puzzles::snapshot-2026-09-18","subject":{"source_id":"gpt-6-astra_max","name":"gpt-6-astra_max","model_id":"gpt-6-astra::max","variant":null,"harness":null},"value":0.84,"unit":"fraction","basis":"measured","source":{"url":"https://epoch.ai/data/benchmark_data.zip","retrieved_at":"2026-09-18T23:49:00Z","published_at":null,"file":"data/raw/benchmarks/daily-evidence/2026-09-18-epoch-hub/epoch-mystery_game_puzzles.csv.gz","sha256":"8915ec1884e41ae8b8e677a826d238b61bbada2d3c9e2c8e74f8096260a8f367","locator":"csv; source row 6; gpt-6-astra_max; field Best score (across scorers)"},"protocol":"Mystery Game Puzzles run by Epoch AI (inspect-ai): 100 programmatic mid-game positions of an undisclosed game, one best move each; value = Best score (across scorers). mean_score, stderr, Started at and the Epoch run id stay in the protocol; CC BY 4.0, attribute Epoch AI.; mean_score=0.84; stderr=0.036845294917747066; Started at=2026-08-30T14:57:43.000Z; id=FA7pM9hKdkgTXmGHptcno7","comparison_key":null,"join_note":"Reviewed identity map 2026-09-16: label states model gpt-6-astra and effort max; exact catalog configuration gpt-6-astra::max"}],"divergences":[],"cell":null,"coverage":{"model":null,"benchmark":{"total_models":844,"available":38,"unknown":806,"not_tested":0,"not_published":0,"source_unreachable":0,"contested":0,"measured":38,"self_reported":0,"observations":128,"unmatched_observations":90}},"collection":{"benchmark_id":"mystery-game-puzzles::snapshot-2026-09-18","status":"collected","source_url":"https://epoch.ai/data/benchmark_data.zip","reason":"128 source results parsed; configurations remain separate; unmatched model identities are retained."},"coverage_note":"Denominators are catalog model configurations × versioned registry entries; capability_available excludes Efficiency (cost) boards. Basis counts count covered cells, not runs; measured and self_reported may overlap. Unmatched source identities remain in observations and never inflate catalog coverage."}