Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
•
470154a
1
Parent(s):
9ab0b0c
Add anaphor_number_agreement data files
Browse files- README.md +7 -3
- anaphor_number_agreement/train-00000-of-00001.parquet +3 -0
- dataset_infos.json +6 -24
README.md
CHANGED
@@ -100,10 +100,10 @@ dataset_info:
|
|
100 |
dtype: int32
|
101 |
splits:
|
102 |
- name: train
|
103 |
-
num_bytes:
|
104 |
num_examples: 1000
|
105 |
-
download_size:
|
106 |
-
dataset_size:
|
107 |
- config_name: animate_subject_passive
|
108 |
features:
|
109 |
- name: sentence_good
|
@@ -1905,6 +1905,10 @@ configs:
|
|
1905 |
data_files:
|
1906 |
- split: train
|
1907 |
path: anaphor_gender_agreement/train-*
|
|
|
|
|
|
|
|
|
1908 |
---
|
1909 |
|
1910 |
# Dataset Card for "blimp"
|
|
|
100 |
dtype: int32
|
101 |
splits:
|
102 |
- name: train
|
103 |
+
num_bytes: 139879
|
104 |
num_examples: 1000
|
105 |
+
download_size: 41547
|
106 |
+
dataset_size: 139879
|
107 |
- config_name: animate_subject_passive
|
108 |
features:
|
109 |
- name: sentence_good
|
|
|
1905 |
data_files:
|
1906 |
- split: train
|
1907 |
path: anaphor_gender_agreement/train-*
|
1908 |
+
- config_name: anaphor_number_agreement
|
1909 |
+
data_files:
|
1910 |
+
- split: train
|
1911 |
+
path: anaphor_number_agreement/train-*
|
1912 |
---
|
1913 |
|
1914 |
# Dataset Card for "blimp"
|
anaphor_number_agreement/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5dc16d19b9b3694589a11a66c7d0d0ca5707d3762a45d8cf0306133151575784
|
3 |
+
size 41547
|
dataset_infos.json
CHANGED
@@ -143,62 +143,50 @@
|
|
143 |
"features": {
|
144 |
"sentence_good": {
|
145 |
"dtype": "string",
|
146 |
-
"id": null,
|
147 |
"_type": "Value"
|
148 |
},
|
149 |
"sentence_bad": {
|
150 |
"dtype": "string",
|
151 |
-
"id": null,
|
152 |
"_type": "Value"
|
153 |
},
|
154 |
"field": {
|
155 |
"dtype": "string",
|
156 |
-
"id": null,
|
157 |
"_type": "Value"
|
158 |
},
|
159 |
"linguistics_term": {
|
160 |
"dtype": "string",
|
161 |
-
"id": null,
|
162 |
"_type": "Value"
|
163 |
},
|
164 |
"UID": {
|
165 |
"dtype": "string",
|
166 |
-
"id": null,
|
167 |
"_type": "Value"
|
168 |
},
|
169 |
"simple_LM_method": {
|
170 |
"dtype": "bool",
|
171 |
-
"id": null,
|
172 |
"_type": "Value"
|
173 |
},
|
174 |
"one_prefix_method": {
|
175 |
"dtype": "bool",
|
176 |
-
"id": null,
|
177 |
"_type": "Value"
|
178 |
},
|
179 |
"two_prefix_method": {
|
180 |
"dtype": "bool",
|
181 |
-
"id": null,
|
182 |
"_type": "Value"
|
183 |
},
|
184 |
"lexically_identical": {
|
185 |
"dtype": "bool",
|
186 |
-
"id": null,
|
187 |
"_type": "Value"
|
188 |
},
|
189 |
"pair_id": {
|
190 |
"dtype": "int32",
|
191 |
-
"id": null,
|
192 |
"_type": "Value"
|
193 |
}
|
194 |
},
|
195 |
-
"supervised_keys": null,
|
196 |
"builder_name": "blimp",
|
|
|
197 |
"config_name": "anaphor_number_agreement",
|
198 |
"version": {
|
199 |
"version_str": "0.1.0",
|
200 |
-
"description": null,
|
201 |
-
"datasets_version_to_prepare": null,
|
202 |
"major": 0,
|
203 |
"minor": 1,
|
204 |
"patch": 0
|
@@ -206,20 +194,14 @@
|
|
206 |
"splits": {
|
207 |
"train": {
|
208 |
"name": "train",
|
209 |
-
"num_bytes":
|
210 |
"num_examples": 1000,
|
211 |
-
"dataset_name":
|
212 |
-
}
|
213 |
-
},
|
214 |
-
"download_checksums": {
|
215 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/anaphor_number_agreement.jsonl": {
|
216 |
-
"num_bytes": 450861,
|
217 |
-
"checksum": "05c16b8bfe4feb5b3552dc2c53d1742ce0c91d9a1c896684335bdddd4373cc9b"
|
218 |
}
|
219 |
},
|
220 |
-
"download_size":
|
221 |
-
"dataset_size":
|
222 |
-
"size_in_bytes":
|
223 |
},
|
224 |
"animate_subject_passive": {
|
225 |
"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",
|
|
|
143 |
"features": {
|
144 |
"sentence_good": {
|
145 |
"dtype": "string",
|
|
|
146 |
"_type": "Value"
|
147 |
},
|
148 |
"sentence_bad": {
|
149 |
"dtype": "string",
|
|
|
150 |
"_type": "Value"
|
151 |
},
|
152 |
"field": {
|
153 |
"dtype": "string",
|
|
|
154 |
"_type": "Value"
|
155 |
},
|
156 |
"linguistics_term": {
|
157 |
"dtype": "string",
|
|
|
158 |
"_type": "Value"
|
159 |
},
|
160 |
"UID": {
|
161 |
"dtype": "string",
|
|
|
162 |
"_type": "Value"
|
163 |
},
|
164 |
"simple_LM_method": {
|
165 |
"dtype": "bool",
|
|
|
166 |
"_type": "Value"
|
167 |
},
|
168 |
"one_prefix_method": {
|
169 |
"dtype": "bool",
|
|
|
170 |
"_type": "Value"
|
171 |
},
|
172 |
"two_prefix_method": {
|
173 |
"dtype": "bool",
|
|
|
174 |
"_type": "Value"
|
175 |
},
|
176 |
"lexically_identical": {
|
177 |
"dtype": "bool",
|
|
|
178 |
"_type": "Value"
|
179 |
},
|
180 |
"pair_id": {
|
181 |
"dtype": "int32",
|
|
|
182 |
"_type": "Value"
|
183 |
}
|
184 |
},
|
|
|
185 |
"builder_name": "blimp",
|
186 |
+
"dataset_name": "blimp",
|
187 |
"config_name": "anaphor_number_agreement",
|
188 |
"version": {
|
189 |
"version_str": "0.1.0",
|
|
|
|
|
190 |
"major": 0,
|
191 |
"minor": 1,
|
192 |
"patch": 0
|
|
|
194 |
"splits": {
|
195 |
"train": {
|
196 |
"name": "train",
|
197 |
+
"num_bytes": 139879,
|
198 |
"num_examples": 1000,
|
199 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
200 |
}
|
201 |
},
|
202 |
+
"download_size": 41547,
|
203 |
+
"dataset_size": 139879,
|
204 |
+
"size_in_bytes": 181426
|
205 |
},
|
206 |
"animate_subject_passive": {
|
207 |
"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",
|