Attend the Women in ML Symposium on December 7 Register now

hendrycks_test

References:

abstract_algebra

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/abstract_algebra')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

anatomy

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/anatomy')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 135
'validation' 14
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

astronomy

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/astronomy')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 152
'validation' 16
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

business_ethics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/business_ethics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

clinical_knowledge

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/clinical_knowledge')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 265
'validation' 29
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

college_biology

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/college_biology')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 144
'validation' 16
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

college_chemistry

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/college_chemistry')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 8
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

college_computer_science

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/college_computer_science')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

college_mathematics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/college_mathematics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

college_medicine

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/college_medicine')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 173
'validation' 22
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

college_physics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/college_physics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 102
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

computer_security

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/computer_security')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

conceptual_physics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/conceptual_physics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 235
'validation' 26
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

econometrics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/econometrics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 114
'validation' 12
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

electrical_engineering

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/electrical_engineering')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 145
'validation' 16
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

elementary_mathematics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/elementary_mathematics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 378
'validation' 41
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

formal_logic

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/formal_logic')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 126
'validation' 14
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

global_facts

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/global_facts')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 10
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_biology

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_biology')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 310
'validation' 32
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_chemistry

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_chemistry')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 203
'validation' 22
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_computer_science

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_computer_science')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 9
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_european_history

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_european_history')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 165
'validation' 18
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_geography

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_geography')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 198
'validation' 22
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_government_and_politics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_government_and_politics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 193
'validation' 21
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_macroeconomics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_macroeconomics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 390
'validation' 43
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_mathematics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_mathematics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 270
'validation' 29
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_microeconomics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_microeconomics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 238
'validation' 26
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_physics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_physics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 151
'validation' 17
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_psychology

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_psychology')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 545
'validation' 60
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_statistics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_statistics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 216
'validation' 23
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_us_history

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_us_history')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 204
'validation' 22
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

high_school_world_history

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/high_school_world_history')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 237
'validation' 26
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

human_aging

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/human_aging')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 223
'validation' 23
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

human_sexuality

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/human_sexuality')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 131
'validation' 12
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

international_law

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/international_law')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 121
'validation' 13
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

jurisprudence

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/jurisprudence')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 108
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

logical_fallacies

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/logical_fallacies')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 163
'validation' 18
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

machine_learning

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/machine_learning')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 112
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

management

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/management')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 103
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

marketing

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/marketing')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 234
'validation' 25
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

medical_genetics

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/medical_genetics')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

miscellaneous

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/miscellaneous')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 783
'validation' 86
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

moral_disputes

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/moral_disputes')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 346
'validation' 38
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

moral_scenarios

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/moral_scenarios')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 895
'validation' 100
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

nutrition

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/nutrition')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 306
'validation' 33
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

philosophy

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/philosophy')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 311
'validation' 34
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

prehistory

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/prehistory')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 324
'validation' 35
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

professional_accounting

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/professional_accounting')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 282
'validation' 31
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

professional_law

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/professional_law')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 1534
'validation' 170
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

professional_medicine

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/professional_medicine')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 272
'validation' 31
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

professional_psychology

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/professional_psychology')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 612
'validation' 69
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

public_relations

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/public_relations')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 110
'validation' 12
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

security_studies

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/security_studies')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 245
'validation' 27
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

sociology

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/sociology')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 201
'validation' 22
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

us_foreign_policy

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/us_foreign_policy')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 100
'validation' 11
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

virology

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/virology')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 166
'validation' 18
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}

world_religions

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:hendrycks_test/world_religions')
  • Description:
This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge, covering 57 tasks including elementary mathematics, US history, computer science, law, and more.
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'auxiliary_train' 99842
'dev' 5
'test' 171
'validation' 19
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "choices": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "answer": {
        "num_classes": 4,
        "names": [
            "A",
            "B",
            "C",
            "D"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}