Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
·
e7a8756
1
Parent(s):
fca9ef9
Add tough_vs_raising_1 data files
Browse files- README.md +7 -3
- dataset_infos.json +6 -24
- tough_vs_raising_1/train-00000-of-00001.parquet +3 -0
README.md
CHANGED
@@ -1612,10 +1612,10 @@ dataset_info:
|
|
1612 |
dtype: int32
|
1613 |
splits:
|
1614 |
- name: train
|
1615 |
-
num_bytes:
|
1616 |
num_examples: 1000
|
1617 |
-
download_size:
|
1618 |
-
dataset_size:
|
1619 |
- config_name: tough_vs_raising_2
|
1620 |
features:
|
1621 |
- name: sentence_good
|
@@ -2121,6 +2121,10 @@ configs:
|
|
2121 |
data_files:
|
2122 |
- split: train
|
2123 |
path: superlative_quantifiers_2/train-*
|
|
|
|
|
|
|
|
|
2124 |
---
|
2125 |
|
2126 |
# Dataset Card for "blimp"
|
|
|
1612 |
dtype: int32
|
1613 |
splits:
|
1614 |
- name: train
|
1615 |
+
num_bytes: 148636
|
1616 |
num_examples: 1000
|
1617 |
+
download_size: 44779
|
1618 |
+
dataset_size: 148636
|
1619 |
- config_name: tough_vs_raising_2
|
1620 |
features:
|
1621 |
- name: sentence_good
|
|
|
2121 |
data_files:
|
2122 |
- split: train
|
2123 |
path: superlative_quantifiers_2/train-*
|
2124 |
+
- config_name: tough_vs_raising_1
|
2125 |
+
data_files:
|
2126 |
+
- split: train
|
2127 |
+
path: tough_vs_raising_1/train-*
|
2128 |
---
|
2129 |
|
2130 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -3815,62 +3815,50 @@
|
|
3815 |
"features": {
|
3816 |
"sentence_good": {
|
3817 |
"dtype": "string",
|
3818 |
-
"id": null,
|
3819 |
"_type": "Value"
|
3820 |
},
|
3821 |
"sentence_bad": {
|
3822 |
"dtype": "string",
|
3823 |
-
"id": null,
|
3824 |
"_type": "Value"
|
3825 |
},
|
3826 |
"field": {
|
3827 |
"dtype": "string",
|
3828 |
-
"id": null,
|
3829 |
"_type": "Value"
|
3830 |
},
|
3831 |
"linguistics_term": {
|
3832 |
"dtype": "string",
|
3833 |
-
"id": null,
|
3834 |
"_type": "Value"
|
3835 |
},
|
3836 |
"UID": {
|
3837 |
"dtype": "string",
|
3838 |
-
"id": null,
|
3839 |
"_type": "Value"
|
3840 |
},
|
3841 |
"simple_LM_method": {
|
3842 |
"dtype": "bool",
|
3843 |
-
"id": null,
|
3844 |
"_type": "Value"
|
3845 |
},
|
3846 |
"one_prefix_method": {
|
3847 |
"dtype": "bool",
|
3848 |
-
"id": null,
|
3849 |
"_type": "Value"
|
3850 |
},
|
3851 |
"two_prefix_method": {
|
3852 |
"dtype": "bool",
|
3853 |
-
"id": null,
|
3854 |
"_type": "Value"
|
3855 |
},
|
3856 |
"lexically_identical": {
|
3857 |
"dtype": "bool",
|
3858 |
-
"id": null,
|
3859 |
"_type": "Value"
|
3860 |
},
|
3861 |
"pair_id": {
|
3862 |
"dtype": "int32",
|
3863 |
-
"id": null,
|
3864 |
"_type": "Value"
|
3865 |
}
|
3866 |
},
|
3867 |
-
"supervised_keys": null,
|
3868 |
"builder_name": "blimp",
|
|
|
3869 |
"config_name": "tough_vs_raising_1",
|
3870 |
"version": {
|
3871 |
"version_str": "0.1.0",
|
3872 |
-
"description": null,
|
3873 |
-
"datasets_version_to_prepare": null,
|
3874 |
"major": 0,
|
3875 |
"minor": 1,
|
3876 |
"patch": 0
|
@@ -3878,20 +3866,14 @@
|
|
3878 |
"splits": {
|
3879 |
"train": {
|
3880 |
"name": "train",
|
3881 |
-
"num_bytes":
|
3882 |
"num_examples": 1000,
|
3883 |
-
"dataset_name":
|
3884 |
-
}
|
3885 |
-
},
|
3886 |
-
"download_checksums": {
|
3887 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/tough_vs_raising_1.jsonl": {
|
3888 |
-
"num_bytes": 343026,
|
3889 |
-
"checksum": "8eb3a1ecbdbd969c5b77e2f85b12a39d33723bdc58d673e5067bb456da91d0e5"
|
3890 |
}
|
3891 |
},
|
3892 |
-
"download_size":
|
3893 |
-
"dataset_size":
|
3894 |
-
"size_in_bytes":
|
3895 |
},
|
3896 |
"tough_vs_raising_2": {
|
3897 |
"description": "\nBLiMP is a challenge set for evaluating what language models (LMs) know about\nmajor grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each\ncontaining 1000 minimal pairs isolating specific contrasts in syntax,\nmorphology, or semantics. The data is automatically generated according to\nexpert-crafted grammars.\n",
|
|
|
3815 |
"features": {
|
3816 |
"sentence_good": {
|
3817 |
"dtype": "string",
|
|
|
3818 |
"_type": "Value"
|
3819 |
},
|
3820 |
"sentence_bad": {
|
3821 |
"dtype": "string",
|
|
|
3822 |
"_type": "Value"
|
3823 |
},
|
3824 |
"field": {
|
3825 |
"dtype": "string",
|
|
|
3826 |
"_type": "Value"
|
3827 |
},
|
3828 |
"linguistics_term": {
|
3829 |
"dtype": "string",
|
|
|
3830 |
"_type": "Value"
|
3831 |
},
|
3832 |
"UID": {
|
3833 |
"dtype": "string",
|
|
|
3834 |
"_type": "Value"
|
3835 |
},
|
3836 |
"simple_LM_method": {
|
3837 |
"dtype": "bool",
|
|
|
3838 |
"_type": "Value"
|
3839 |
},
|
3840 |
"one_prefix_method": {
|
3841 |
"dtype": "bool",
|
|
|
3842 |
"_type": "Value"
|
3843 |
},
|
3844 |
"two_prefix_method": {
|
3845 |
"dtype": "bool",
|
|
|
3846 |
"_type": "Value"
|
3847 |
},
|
3848 |
"lexically_identical": {
|
3849 |
"dtype": "bool",
|
|
|
3850 |
"_type": "Value"
|
3851 |
},
|
3852 |
"pair_id": {
|
3853 |
"dtype": "int32",
|
|
|
3854 |
"_type": "Value"
|
3855 |
}
|
3856 |
},
|
|
|
3857 |
"builder_name": "blimp",
|
3858 |
+
"dataset_name": "blimp",
|
3859 |
"config_name": "tough_vs_raising_1",
|
3860 |
"version": {
|
3861 |
"version_str": "0.1.0",
|
|
|
|
|
3862 |
"major": 0,
|
3863 |
"minor": 1,
|
3864 |
"patch": 0
|
|
|
3866 |
"splits": {
|
3867 |
"train": {
|
3868 |
"name": "train",
|
3869 |
+
"num_bytes": 148636,
|
3870 |
"num_examples": 1000,
|
3871 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
3872 |
}
|
3873 |
},
|
3874 |
+
"download_size": 44779,
|
3875 |
+
"dataset_size": 148636,
|
3876 |
+
"size_in_bytes": 193415
|
3877 |
},
|
3878 |
"tough_vs_raising_2": {
|
3879 |
"description": "\nBLiMP is a challenge set for evaluating what language models (LMs) know about\nmajor grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each\ncontaining 1000 minimal pairs isolating specific contrasts in syntax,\nmorphology, or semantics. The data is automatically generated according to\nexpert-crafted grammars.\n",
|
tough_vs_raising_1/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:defbb5c0cb9b3d4b4b5f1571e64df8ba37428619bb4943d4d228ad5caf83f4d7
|
3 |
+
size 44779
|