Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
•
9a8f400
1
Parent(s):
5b76995
Add wh_questions_subject_gap_long_distance data files
Browse files
README.md
CHANGED
@@ -1780,10 +1780,10 @@ dataset_info:
|
|
1780 |
dtype: int32
|
1781 |
splits:
|
1782 |
- name: train
|
1783 |
-
num_bytes:
|
1784 |
num_examples: 1000
|
1785 |
-
download_size:
|
1786 |
-
dataset_size:
|
1787 |
- config_name: wh_vs_that_no_gap
|
1788 |
features:
|
1789 |
- name: sentence_good
|
@@ -2145,6 +2145,10 @@ configs:
|
|
2145 |
data_files:
|
2146 |
- split: train
|
2147 |
path: wh_questions_subject_gap/train-*
|
|
|
|
|
|
|
|
|
2148 |
---
|
2149 |
|
2150 |
# Dataset Card for "blimp"
|
|
|
1780 |
dtype: int32
|
1781 |
splits:
|
1782 |
- name: train
|
1783 |
+
num_bytes: 268270
|
1784 |
num_examples: 1000
|
1785 |
+
download_size: 98913
|
1786 |
+
dataset_size: 268270
|
1787 |
- config_name: wh_vs_that_no_gap
|
1788 |
features:
|
1789 |
- name: sentence_good
|
|
|
2145 |
data_files:
|
2146 |
- split: train
|
2147 |
path: wh_questions_subject_gap/train-*
|
2148 |
+
- config_name: wh_questions_subject_gap_long_distance
|
2149 |
+
data_files:
|
2150 |
+
- split: train
|
2151 |
+
path: wh_questions_subject_gap_long_distance/train-*
|
2152 |
---
|
2153 |
|
2154 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -4223,62 +4223,50 @@
|
|
4223 |
"features": {
|
4224 |
"sentence_good": {
|
4225 |
"dtype": "string",
|
4226 |
-
"id": null,
|
4227 |
"_type": "Value"
|
4228 |
},
|
4229 |
"sentence_bad": {
|
4230 |
"dtype": "string",
|
4231 |
-
"id": null,
|
4232 |
"_type": "Value"
|
4233 |
},
|
4234 |
"field": {
|
4235 |
"dtype": "string",
|
4236 |
-
"id": null,
|
4237 |
"_type": "Value"
|
4238 |
},
|
4239 |
"linguistics_term": {
|
4240 |
"dtype": "string",
|
4241 |
-
"id": null,
|
4242 |
"_type": "Value"
|
4243 |
},
|
4244 |
"UID": {
|
4245 |
"dtype": "string",
|
4246 |
-
"id": null,
|
4247 |
"_type": "Value"
|
4248 |
},
|
4249 |
"simple_LM_method": {
|
4250 |
"dtype": "bool",
|
4251 |
-
"id": null,
|
4252 |
"_type": "Value"
|
4253 |
},
|
4254 |
"one_prefix_method": {
|
4255 |
"dtype": "bool",
|
4256 |
-
"id": null,
|
4257 |
"_type": "Value"
|
4258 |
},
|
4259 |
"two_prefix_method": {
|
4260 |
"dtype": "bool",
|
4261 |
-
"id": null,
|
4262 |
"_type": "Value"
|
4263 |
},
|
4264 |
"lexically_identical": {
|
4265 |
"dtype": "bool",
|
4266 |
-
"id": null,
|
4267 |
"_type": "Value"
|
4268 |
},
|
4269 |
"pair_id": {
|
4270 |
"dtype": "int32",
|
4271 |
-
"id": null,
|
4272 |
"_type": "Value"
|
4273 |
}
|
4274 |
},
|
4275 |
-
"supervised_keys": null,
|
4276 |
"builder_name": "blimp",
|
|
|
4277 |
"config_name": "wh_questions_subject_gap_long_distance",
|
4278 |
"version": {
|
4279 |
"version_str": "0.1.0",
|
4280 |
-
"description": null,
|
4281 |
-
"datasets_version_to_prepare": null,
|
4282 |
"major": 0,
|
4283 |
"minor": 1,
|
4284 |
"patch": 0
|
@@ -4286,20 +4274,14 @@
|
|
4286 |
"splits": {
|
4287 |
"train": {
|
4288 |
"name": "train",
|
4289 |
-
"num_bytes":
|
4290 |
"num_examples": 1000,
|
4291 |
-
"dataset_name":
|
4292 |
-
}
|
4293 |
-
},
|
4294 |
-
"download_checksums": {
|
4295 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/wh_questions_subject_gap_long_distance.jsonl": {
|
4296 |
-
"num_bytes": 462660,
|
4297 |
-
"checksum": "985bae202dbc5566b9b5c51bd77dd9d1dbcd87e0f38e008c34a11d7aa07db788"
|
4298 |
}
|
4299 |
},
|
4300 |
-
"download_size":
|
4301 |
-
"dataset_size":
|
4302 |
-
"size_in_bytes":
|
4303 |
},
|
4304 |
"wh_vs_that_no_gap": {
|
4305 |
"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",
|
|
|
4223 |
"features": {
|
4224 |
"sentence_good": {
|
4225 |
"dtype": "string",
|
|
|
4226 |
"_type": "Value"
|
4227 |
},
|
4228 |
"sentence_bad": {
|
4229 |
"dtype": "string",
|
|
|
4230 |
"_type": "Value"
|
4231 |
},
|
4232 |
"field": {
|
4233 |
"dtype": "string",
|
|
|
4234 |
"_type": "Value"
|
4235 |
},
|
4236 |
"linguistics_term": {
|
4237 |
"dtype": "string",
|
|
|
4238 |
"_type": "Value"
|
4239 |
},
|
4240 |
"UID": {
|
4241 |
"dtype": "string",
|
|
|
4242 |
"_type": "Value"
|
4243 |
},
|
4244 |
"simple_LM_method": {
|
4245 |
"dtype": "bool",
|
|
|
4246 |
"_type": "Value"
|
4247 |
},
|
4248 |
"one_prefix_method": {
|
4249 |
"dtype": "bool",
|
|
|
4250 |
"_type": "Value"
|
4251 |
},
|
4252 |
"two_prefix_method": {
|
4253 |
"dtype": "bool",
|
|
|
4254 |
"_type": "Value"
|
4255 |
},
|
4256 |
"lexically_identical": {
|
4257 |
"dtype": "bool",
|
|
|
4258 |
"_type": "Value"
|
4259 |
},
|
4260 |
"pair_id": {
|
4261 |
"dtype": "int32",
|
|
|
4262 |
"_type": "Value"
|
4263 |
}
|
4264 |
},
|
|
|
4265 |
"builder_name": "blimp",
|
4266 |
+
"dataset_name": "blimp",
|
4267 |
"config_name": "wh_questions_subject_gap_long_distance",
|
4268 |
"version": {
|
4269 |
"version_str": "0.1.0",
|
|
|
|
|
4270 |
"major": 0,
|
4271 |
"minor": 1,
|
4272 |
"patch": 0
|
|
|
4274 |
"splits": {
|
4275 |
"train": {
|
4276 |
"name": "train",
|
4277 |
+
"num_bytes": 268270,
|
4278 |
"num_examples": 1000,
|
4279 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
4280 |
}
|
4281 |
},
|
4282 |
+
"download_size": 98913,
|
4283 |
+
"dataset_size": 268270,
|
4284 |
+
"size_in_bytes": 367183
|
4285 |
},
|
4286 |
"wh_vs_that_no_gap": {
|
4287 |
"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",
|
wh_questions_subject_gap_long_distance/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ca30473b9569b256cc7707d02668db436d097c810466c8cb60b1b7315ddb10e9
|
3 |
+
size 98913
|