cluster_loras / 20 /mmlu /metrics.json
Muqeeth's picture
Upload folder using huggingface_hub
9224edf verified
{"average_acc": 0.42351516877937617, "subcat_acc": {"math": 0.2781954887218045, "health": 0.44390243902439025, "physics": 0.3359375, "business": 0.5560640732265446, "biology": 0.43832599118942733, "chemistry": 0.3201320132013201, "computer science": 0.3932038834951456, "economics": 0.38544474393531, "engineering": 0.4, "philosophy": 0.36083499005964215, "other": 0.4875536480686695, "history": 0.5279569892473118, "geography": 0.47474747474747475, "politics": 0.5385802469135802, "psychology": 0.4736387208297321, "culture": 0.6114457831325302, "law": 0.3879750425411231}, "cat_acc": {"STEM": 0.34029158383035124, "humanities": 0.40403825717322, "social sciences": 0.4809879753006175, "other (business, health, misc.)": 0.47470697100555215}, "subjects": {"abstract_algebra": 0.25, "anatomy": 0.42962962962962964, "astronomy": 0.4342105263157895, "business_ethics": 0.43, "clinical_knowledge": 0.47924528301886793, "college_biology": 0.4305555555555556, "college_chemistry": 0.36, "college_computer_science": 0.37, "college_mathematics": 0.34, "college_medicine": 0.45664739884393063, "college_physics": 0.21568627450980393, "computer_security": 0.46, "conceptual_physics": 0.3446808510638298, "econometrics": 0.24561403508771928, "electrical_engineering": 0.4, "elementary_mathematics": 0.24867724867724866, "formal_logic": 0.38095238095238093, "global_facts": 0.2, "high_school_biology": 0.44193548387096776, "high_school_chemistry": 0.30049261083743845, "high_school_computer_science": 0.35, "high_school_european_history": 0.6060606060606061, "high_school_geography": 0.47474747474747475, "high_school_government_and_politics": 0.5854922279792746, "high_school_macroeconomics": 0.41794871794871796, "high_school_mathematics": 0.25555555555555554, "high_school_microeconomics": 0.39915966386554624, "high_school_physics": 0.304635761589404, "high_school_psychology": 0.5266055045871559, "high_school_statistics": 0.3425925925925926, "high_school_us_history": 0.5294117647058824, "high_school_world_history": 0.5780590717299579, "human_aging": 0.4170403587443946, "human_sexuality": 0.5648854961832062, "international_law": 0.5867768595041323, "jurisprudence": 0.49074074074074076, "logical_fallacies": 0.49693251533742333, "machine_learning": 0.39285714285714285, "management": 0.44660194174757284, "marketing": 0.6581196581196581, "medical_genetics": 0.47, "miscellaneous": 0.5708812260536399, "moral_disputes": 0.3815028901734104, "moral_scenarios": 0.23910614525139665, "nutrition": 0.434640522875817, "philosophy": 0.4758842443729904, "prehistory": 0.4506172839506173, "professional_accounting": 0.35815602836879434, "professional_law": 0.3650586701434159, "professional_medicine": 0.4411764705882353, "professional_psychology": 0.4264705882352941, "public_relations": 0.44545454545454544, "security_studies": 0.5061224489795918, "sociology": 0.6417910447761194, "us_foreign_policy": 0.63, "virology": 0.42771084337349397, "world_religions": 0.6023391812865497}}