Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
•
6d579cb
1
Parent(s):
d458d7f
Add principle_A_c_command data files
Browse files- README.md +7 -3
- dataset_infos.json +6 -24
- principle_A_c_command/train-00000-of-00001.parquet +3 -0
README.md
CHANGED
@@ -1220,10 +1220,10 @@ dataset_info:
|
|
1220 |
dtype: int32
|
1221 |
splits:
|
1222 |
- name: train
|
1223 |
-
num_bytes:
|
1224 |
num_examples: 1000
|
1225 |
-
download_size:
|
1226 |
-
dataset_size:
|
1227 |
- config_name: principle_A_case_1
|
1228 |
features:
|
1229 |
- name: sentence_good
|
@@ -2065,6 +2065,10 @@ configs:
|
|
2065 |
data_files:
|
2066 |
- split: train
|
2067 |
path: passive_2/train-*
|
|
|
|
|
|
|
|
|
2068 |
---
|
2069 |
|
2070 |
# Dataset Card for "blimp"
|
|
|
1220 |
dtype: int32
|
1221 |
splits:
|
1222 |
- name: train
|
1223 |
+
num_bytes: 188490
|
1224 |
num_examples: 1000
|
1225 |
+
download_size: 67867
|
1226 |
+
dataset_size: 188490
|
1227 |
- config_name: principle_A_case_1
|
1228 |
features:
|
1229 |
- name: sentence_good
|
|
|
2065 |
data_files:
|
2066 |
- split: train
|
2067 |
path: passive_2/train-*
|
2068 |
+
- config_name: principle_A_c_command
|
2069 |
+
data_files:
|
2070 |
+
- split: train
|
2071 |
+
path: principle_A_c_command/train-*
|
2072 |
---
|
2073 |
|
2074 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -2863,62 +2863,50 @@
|
|
2863 |
"features": {
|
2864 |
"sentence_good": {
|
2865 |
"dtype": "string",
|
2866 |
-
"id": null,
|
2867 |
"_type": "Value"
|
2868 |
},
|
2869 |
"sentence_bad": {
|
2870 |
"dtype": "string",
|
2871 |
-
"id": null,
|
2872 |
"_type": "Value"
|
2873 |
},
|
2874 |
"field": {
|
2875 |
"dtype": "string",
|
2876 |
-
"id": null,
|
2877 |
"_type": "Value"
|
2878 |
},
|
2879 |
"linguistics_term": {
|
2880 |
"dtype": "string",
|
2881 |
-
"id": null,
|
2882 |
"_type": "Value"
|
2883 |
},
|
2884 |
"UID": {
|
2885 |
"dtype": "string",
|
2886 |
-
"id": null,
|
2887 |
"_type": "Value"
|
2888 |
},
|
2889 |
"simple_LM_method": {
|
2890 |
"dtype": "bool",
|
2891 |
-
"id": null,
|
2892 |
"_type": "Value"
|
2893 |
},
|
2894 |
"one_prefix_method": {
|
2895 |
"dtype": "bool",
|
2896 |
-
"id": null,
|
2897 |
"_type": "Value"
|
2898 |
},
|
2899 |
"two_prefix_method": {
|
2900 |
"dtype": "bool",
|
2901 |
-
"id": null,
|
2902 |
"_type": "Value"
|
2903 |
},
|
2904 |
"lexically_identical": {
|
2905 |
"dtype": "bool",
|
2906 |
-
"id": null,
|
2907 |
"_type": "Value"
|
2908 |
},
|
2909 |
"pair_id": {
|
2910 |
"dtype": "int32",
|
2911 |
-
"id": null,
|
2912 |
"_type": "Value"
|
2913 |
}
|
2914 |
},
|
2915 |
-
"supervised_keys": null,
|
2916 |
"builder_name": "blimp",
|
|
|
2917 |
"config_name": "principle_A_c_command",
|
2918 |
"version": {
|
2919 |
"version_str": "0.1.0",
|
2920 |
-
"description": null,
|
2921 |
-
"datasets_version_to_prepare": null,
|
2922 |
"major": 0,
|
2923 |
"minor": 1,
|
2924 |
"patch": 0
|
@@ -2926,20 +2914,14 @@
|
|
2926 |
"splits": {
|
2927 |
"train": {
|
2928 |
"name": "train",
|
2929 |
-
"num_bytes":
|
2930 |
"num_examples": 1000,
|
2931 |
-
"dataset_name":
|
2932 |
-
}
|
2933 |
-
},
|
2934 |
-
"download_checksums": {
|
2935 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/principle_A_c_command.jsonl": {
|
2936 |
-
"num_bytes": 527689,
|
2937 |
-
"checksum": "74c8111666542c797dee1c38c9e432b0b65751dfa0faa51874a990f2cbfc76fc"
|
2938 |
}
|
2939 |
},
|
2940 |
-
"download_size":
|
2941 |
-
"dataset_size":
|
2942 |
-
"size_in_bytes":
|
2943 |
},
|
2944 |
"principle_A_case_1": {
|
2945 |
"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",
|
|
|
2863 |
"features": {
|
2864 |
"sentence_good": {
|
2865 |
"dtype": "string",
|
|
|
2866 |
"_type": "Value"
|
2867 |
},
|
2868 |
"sentence_bad": {
|
2869 |
"dtype": "string",
|
|
|
2870 |
"_type": "Value"
|
2871 |
},
|
2872 |
"field": {
|
2873 |
"dtype": "string",
|
|
|
2874 |
"_type": "Value"
|
2875 |
},
|
2876 |
"linguistics_term": {
|
2877 |
"dtype": "string",
|
|
|
2878 |
"_type": "Value"
|
2879 |
},
|
2880 |
"UID": {
|
2881 |
"dtype": "string",
|
|
|
2882 |
"_type": "Value"
|
2883 |
},
|
2884 |
"simple_LM_method": {
|
2885 |
"dtype": "bool",
|
|
|
2886 |
"_type": "Value"
|
2887 |
},
|
2888 |
"one_prefix_method": {
|
2889 |
"dtype": "bool",
|
|
|
2890 |
"_type": "Value"
|
2891 |
},
|
2892 |
"two_prefix_method": {
|
2893 |
"dtype": "bool",
|
|
|
2894 |
"_type": "Value"
|
2895 |
},
|
2896 |
"lexically_identical": {
|
2897 |
"dtype": "bool",
|
|
|
2898 |
"_type": "Value"
|
2899 |
},
|
2900 |
"pair_id": {
|
2901 |
"dtype": "int32",
|
|
|
2902 |
"_type": "Value"
|
2903 |
}
|
2904 |
},
|
|
|
2905 |
"builder_name": "blimp",
|
2906 |
+
"dataset_name": "blimp",
|
2907 |
"config_name": "principle_A_c_command",
|
2908 |
"version": {
|
2909 |
"version_str": "0.1.0",
|
|
|
|
|
2910 |
"major": 0,
|
2911 |
"minor": 1,
|
2912 |
"patch": 0
|
|
|
2914 |
"splits": {
|
2915 |
"train": {
|
2916 |
"name": "train",
|
2917 |
+
"num_bytes": 188490,
|
2918 |
"num_examples": 1000,
|
2919 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
2920 |
}
|
2921 |
},
|
2922 |
+
"download_size": 67867,
|
2923 |
+
"dataset_size": 188490,
|
2924 |
+
"size_in_bytes": 256357
|
2925 |
},
|
2926 |
"principle_A_case_1": {
|
2927 |
"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",
|
principle_A_c_command/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6d18116ede2c686232348552e632f04be358cc62ce1a8c3d65e32263c625c887
|
3 |
+
size 67867
|