dirigível

Referências:

• Código
• Abraço

ilha_adjunto

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/adjunct_island')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

anáforo_gênero_acordo

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/anaphor_gender_agreement')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

anáforo_número_acordo

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/anaphor_number_agreement')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

animate_subject_passive

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/animate_subject_passive')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

animate_subject_trans

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/animate_subject_trans')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

causador

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/causative')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

complexo_NP_ilha

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/complex_NP_island')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

coordenada_estrutura_constraint_complex_left_branch

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/coordinate_structure_constraint_complex_left_branch')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

coordenada_estrutura_constraint_object_extraction

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/coordinate_structure_constraint_object_extraction')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

determinante_noun_agreement_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/determiner_noun_agreement_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

determinante_noun_agreement_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/determiner_noun_agreement_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

determinante_noun_agreement_irregular_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/determiner_noun_agreement_irregular_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

determinante_noun_agreement_irregular_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/determiner_noun_agreement_irregular_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

determinante_noun_agreement_with_adj_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/determiner_noun_agreement_with_adj_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

determiner_noun_agreement_with_adj_irregular_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/determiner_noun_agreement_with_adj_irregular_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

determiner_noun_agreement_with_adj_irregular_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/determiner_noun_agreement_with_adj_irregular_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

determinante_noun_agreement_with_adjective_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/determiner_noun_agreement_with_adjective_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

distrator_agreement_relational_noun

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/distractor_agreement_relational_noun')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

distrator_acordo_relativo_cláusula

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/distractor_agreement_relative_clause')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

drop_argument

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/drop_argument')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

reticências_n_bar_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/ellipsis_n_bar_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

reticências_n_bar_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/ellipsis_n_bar_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

existencial_there_object_raising

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/existential_there_object_raising')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

existencial_there_quantifiers_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/existential_there_quantifiers_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

existencial_there_quantifiers_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/existential_there_quantifiers_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

existencial_there_subject_raising

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/existential_there_subject_raising')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

expletive_it_object_raising

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/expletive_it_object_raising')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

incoativo

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/inchoative')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

intransitivo

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/intransitive')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

irregular_past_participle_adjectives

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/irregular_past_participle_adjectives')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

irregular_past_participle_verbs

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/irregular_past_participle_verbs')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

irregular_plural_subject_verb_agreement_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/irregular_plural_subject_verb_agreement_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

irregular_plural_subject_verb_agreement_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/irregular_plural_subject_verb_agreement_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

left_branch_island_echo_question

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/left_branch_island_echo_question')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

left_branch_island_simple_question

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/left_branch_island_simple_question')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

matrix_question_npi_licensor_present

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/matrix_question_npi_licensor_present')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

npi_presente_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/npi_present_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

npi_presente_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/npi_present_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

only_npi_licenser_present

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/only_npi_licensor_present')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

only_npi_scope

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/only_npi_scope')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

passiva_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/passive_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

passivo_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/passive_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

Principle_A_c_command

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/principle_A_c_command')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

princípio_A_case_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/principle_A_case_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

princípio_A_case_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/principle_A_case_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

princípio_A_domínio_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/principle_A_domain_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

princípio_A_domínio_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/principle_A_domain_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

princípio_A_domínio_3

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/principle_A_domain_3')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

princípio_A_reconstrução

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/principle_A_reconstruction')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

regular_plural_subject_verb_agreement_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/regular_plural_subject_verb_agreement_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

regular_plural_subject_verb_agreement_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/regular_plural_subject_verb_agreement_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

sentential_negation_npi_licensor_present

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/sentential_negation_npi_licensor_present')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

sentential_negation_npi_scope

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/sentential_negation_npi_scope')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

sentential_subject_island

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/sentential_subject_island')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

superlativo_quantifiers_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/superlative_quantifiers_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

superlativo_quantifiers_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/superlative_quantifiers_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

hard_vs_raising_1

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/tough_vs_raising_1')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

hard_vs_raising_2

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/tough_vs_raising_2')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

transitivo

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/transitive')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

wh_island

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/wh_island')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

wh_questions_object_gap

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/wh_questions_object_gap')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

wh_questions_subject_gap

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/wh_questions_subject_gap')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

wh_questions_subject_gap_long_distance

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/wh_questions_subject_gap_long_distance')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

wh_vs_that_no_gap

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/wh_vs_that_no_gap')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

wh_vs_that_no_gap_long_distance

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/wh_vs_that_no_gap_long_distance')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

wh_vs_that_with_gap

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/wh_vs_that_with_gap')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}

wh_vs_that_with_gap_long_distance

Use o seguinte comando para carregar esse conjunto de dados no TFDS:

ds = tfds.load('huggingface:blimp/wh_vs_that_with_gap_long_distance')

• Descrição :

BLiMP is a challenge set for evaluating what language models (LMs) know about
major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each
containing 1000 minimal pairs isolating specific contrasts in syntax,
morphology, or semantics. The data is automatically generated according to
expert-crafted grammars.

• Licença : Nenhuma licença conhecida
• Versão : 0.1.0
• Divisões :

Dividir	Exemplos
'train'	1000

• Características :

{
    "sentence_good": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence_bad": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "field": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "linguistics_term": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "UID": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "simple_LM_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "one_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "two_prefix_method": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "lexically_identical": {
        "dtype": "bool",
        "id": null,
        "_type": "Value"
    },
    "pair_id": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    }
}