Phi 3 mini 4K instruct

Microsoft · microsoft/phi-3-mini-4k-instruct

ProviderMicrosoft
Familyphi
Typellm-code
Statusactive
Released2024-04-22

Reported benchmark scores

These scores carry one collection date for the whole card (2024-07) and a source list rather than a source per score, so they cannot be attributed individually. They are shown as reported and are not verified evidence. Benchmark pages carry the reviewed, dated evidence where it exists.
BenchmarkCatalogue standingScore
ARC-Challengeunassessed63.0
BIG-Bench Hardunassessed55.0
GPQA Diamondunassessed32.3
GSM8Kunassessed74.5
HellaSwagunassessed80.6
IFEvalunassessed54.4
MATH-500unassessed16.4
MMLU: Abstract Algebraunassessed36.0
MMLU: Anatomyunassessed66.7
MMLU: Astronomyunassessed77.0
MMLU: Business Ethicsunassessed68.0
MMLU: Clinical Knowledgeunassessed74.3
MMLU: College Biologyunassessed82.6
MMLU: College Chemistryunassessed47.0
MMLU: College Computer Scienceunassessed55.0
MMLU: College Mathematicsunassessed35.0
MMLU: College Medicineunassessed68.8
MMLU: College Physicsunassessed36.3
MMLU: Computer Securityunassessed78.0
MMLU: Conceptual Physicsunassessed70.6
MMLU: Econometricsunassessed45.6
MMLU: Electrical Engineeringunassessed57.9
MMLU: Elementary Mathematicsunassessed52.6
MMLU: Formal Logicunassessed58.7
MMLU: Global Factsunassessed40.0
MMLU: High School Biologyunassessed83.9
MMLU: High School Chemistryunassessed60.1
MMLU: High School Computer Scienceunassessed73.0
MMLU: High School European Historyunassessed80.6
MMLU: High School Geographyunassessed86.9
MMLU: High School Government and Politicsunassessed90.7
MMLU: High School Macroeconomicsunassessed75.4
MMLU: High School Mathematicsunassessed39.6
MMLU: High School Microeconomicsunassessed83.6
MMLU: High School Physicsunassessed44.4
MMLU: High School Psychologyunassessed88.3
MMLU: High School Statisticsunassessed61.6
MMLU: High School US Historyunassessed79.9
MMLU: High School World Historyunassessed79.3
MMLU: Human Agingunassessed69.1
MMLU: Human Sexualityunassessed76.3
MMLU: International Lawunassessed85.1
MMLU: Jurisprudenceunassessed79.6
MMLU: Logical Fallaciesunassessed80.4
MMLU: Machine Learningunassessed55.4
MMLU: Managementunassessed80.6
MMLU: Marketingunassessed90.2
MMLU: Medical Geneticsunassessed81.0
MMLU: Miscellaneousunassessed82.8
MMLU: Moral Disputesunassessed75.4
MMLU: Moral Scenariosunassessed58.5
MMLU: Nutritionunassessed75.8
MMLU: Philosophyunassessed76.2
MMLU: Prehistoryunassessed78.1
MMLU-Prounassessed40.3
MMLU: Professional Accountingunassessed58.2
MMLU: Professional Lawunassessed51.3
MMLU: Professional Medicineunassessed76.1
MMLU: Professional Psychologyunassessed75.8
MMLU: Public Relationsunassessed73.6
MMLU: Security Studiesunassessed76.7
MMLU: Sociologyunassessed86.1
MMLU: US Foreign Policyunassessed85.0
MMLU: Virologyunassessed49.4
MMLU: World Religionsunassessed83.6
MuSRunassessed42.8
TruthfulQAunassessed59.9
WinoGrandeunassessed72.4

Data

This card as JSON · Edit on GitHub