{ "benchmark": "MMLU finance and business subset", "confidence_interval_95": [ 0.71513558, 0.77929395 ], "dataset": "cais/mmlu", "dataset_revision": "c30699e8356da336a370243923dbaf21066bb9fe", "device": "mps", "dtype": "bfloat16", "evaluated_at": "2026-07-29T10:03:49.030340+00:00", "few_shot": 5, "macro_accuracy": 0.74857143, "micro_accuracy": 0.74857143, "model": ".artifacts/blum-transformers-v1/portable", "model_revision": "ad6f5cec7f729370d2976d8c78983521cb37ca83", "official_leaderboard_result": false, "sample_limit_per_subject": 100, "sample_size": 700, "status": "community_evaluation", "subjects": [ { "accuracy": 0.77, "confidence_interval_95": [ 0.67845617, 0.84156734 ], "correct": 77, "duration_seconds": 285.586, "sample_size": 100, "subject": "business_ethics" }, { "accuracy": 0.65, "confidence_interval_95": [ 0.55254443, 0.73635752 ], "correct": 65, "duration_seconds": 222.028, "sample_size": 100, "subject": "econometrics" }, { "accuracy": 0.7, "confidence_interval_95": [ 0.60415145, 0.78105115 ], "correct": 70, "duration_seconds": 187.56, "sample_size": 100, "subject": "high_school_macroeconomics" }, { "accuracy": 0.87, "confidence_interval_95": [ 0.79019649, 0.92242833 ], "correct": 87, "duration_seconds": 171.766, "sample_size": 100, "subject": "high_school_microeconomics" }, { "accuracy": 0.87, "confidence_interval_95": [ 0.79019649, 0.92242833 ], "correct": 87, "duration_seconds": 101.632, "sample_size": 100, "subject": "management" }, { "accuracy": 0.92, "confidence_interval_95": [ 0.85001892, 0.95890654 ], "correct": 92, "duration_seconds": 194.458, "sample_size": 100, "subject": "marketing" }, { "accuracy": 0.46, "confidence_interval_95": [ 0.36560808, 0.5573514 ], "correct": 46, "duration_seconds": 543.284, "sample_size": 100, "subject": "professional_accounting" } ] }