Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
·
094ce12
1
Parent(s):
0595718
Add principle_A_domain_3 data files
Browse files- README.md +7 -3
- dataset_infos.json +6 -24
- principle_A_domain_3/train-00000-of-00001.parquet +3 -0
README.md
CHANGED
|
@@ -1360,10 +1360,10 @@ dataset_info:
|
|
| 1360 |
dtype: int32
|
| 1361 |
splits:
|
| 1362 |
- name: train
|
| 1363 |
-
num_bytes:
|
| 1364 |
num_examples: 1000
|
| 1365 |
-
download_size:
|
| 1366 |
-
dataset_size:
|
| 1367 |
- config_name: principle_A_reconstruction
|
| 1368 |
features:
|
| 1369 |
- name: sentence_good
|
|
@@ -2085,6 +2085,10 @@ configs:
|
|
| 2085 |
data_files:
|
| 2086 |
- split: train
|
| 2087 |
path: principle_A_domain_2/train-*
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2088 |
---
|
| 2089 |
|
| 2090 |
# Dataset Card for "blimp"
|
|
|
|
| 1360 |
dtype: int32
|
| 1361 |
splits:
|
| 1362 |
- name: train
|
| 1363 |
+
num_bytes: 158998
|
| 1364 |
num_examples: 1000
|
| 1365 |
+
download_size: 52859
|
| 1366 |
+
dataset_size: 158998
|
| 1367 |
- config_name: principle_A_reconstruction
|
| 1368 |
features:
|
| 1369 |
- name: sentence_good
|
|
|
|
| 2085 |
data_files:
|
| 2086 |
- split: train
|
| 2087 |
path: principle_A_domain_2/train-*
|
| 2088 |
+
- config_name: principle_A_domain_3
|
| 2089 |
+
data_files:
|
| 2090 |
+
- split: train
|
| 2091 |
+
path: principle_A_domain_3/train-*
|
| 2092 |
---
|
| 2093 |
|
| 2094 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
|
@@ -3203,62 +3203,50 @@
|
|
| 3203 |
"features": {
|
| 3204 |
"sentence_good": {
|
| 3205 |
"dtype": "string",
|
| 3206 |
-
"id": null,
|
| 3207 |
"_type": "Value"
|
| 3208 |
},
|
| 3209 |
"sentence_bad": {
|
| 3210 |
"dtype": "string",
|
| 3211 |
-
"id": null,
|
| 3212 |
"_type": "Value"
|
| 3213 |
},
|
| 3214 |
"field": {
|
| 3215 |
"dtype": "string",
|
| 3216 |
-
"id": null,
|
| 3217 |
"_type": "Value"
|
| 3218 |
},
|
| 3219 |
"linguistics_term": {
|
| 3220 |
"dtype": "string",
|
| 3221 |
-
"id": null,
|
| 3222 |
"_type": "Value"
|
| 3223 |
},
|
| 3224 |
"UID": {
|
| 3225 |
"dtype": "string",
|
| 3226 |
-
"id": null,
|
| 3227 |
"_type": "Value"
|
| 3228 |
},
|
| 3229 |
"simple_LM_method": {
|
| 3230 |
"dtype": "bool",
|
| 3231 |
-
"id": null,
|
| 3232 |
"_type": "Value"
|
| 3233 |
},
|
| 3234 |
"one_prefix_method": {
|
| 3235 |
"dtype": "bool",
|
| 3236 |
-
"id": null,
|
| 3237 |
"_type": "Value"
|
| 3238 |
},
|
| 3239 |
"two_prefix_method": {
|
| 3240 |
"dtype": "bool",
|
| 3241 |
-
"id": null,
|
| 3242 |
"_type": "Value"
|
| 3243 |
},
|
| 3244 |
"lexically_identical": {
|
| 3245 |
"dtype": "bool",
|
| 3246 |
-
"id": null,
|
| 3247 |
"_type": "Value"
|
| 3248 |
},
|
| 3249 |
"pair_id": {
|
| 3250 |
"dtype": "int32",
|
| 3251 |
-
"id": null,
|
| 3252 |
"_type": "Value"
|
| 3253 |
}
|
| 3254 |
},
|
| 3255 |
-
"supervised_keys": null,
|
| 3256 |
"builder_name": "blimp",
|
|
|
|
| 3257 |
"config_name": "principle_A_domain_3",
|
| 3258 |
"version": {
|
| 3259 |
"version_str": "0.1.0",
|
| 3260 |
-
"description": null,
|
| 3261 |
-
"datasets_version_to_prepare": null,
|
| 3262 |
"major": 0,
|
| 3263 |
"minor": 1,
|
| 3264 |
"patch": 0
|
|
@@ -3266,20 +3254,14 @@
|
|
| 3266 |
"splits": {
|
| 3267 |
"train": {
|
| 3268 |
"name": "train",
|
| 3269 |
-
"num_bytes":
|
| 3270 |
"num_examples": 1000,
|
| 3271 |
-
"dataset_name":
|
| 3272 |
-
}
|
| 3273 |
-
},
|
| 3274 |
-
"download_checksums": {
|
| 3275 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/principle_A_domain_3.jsonl": {
|
| 3276 |
-
"num_bytes": 513886,
|
| 3277 |
-
"checksum": "a0d66b9d8541ab1df08f629161c6c4e8bca3aa2693eaa342edd2097e6b579bdd"
|
| 3278 |
}
|
| 3279 |
},
|
| 3280 |
-
"download_size":
|
| 3281 |
-
"dataset_size":
|
| 3282 |
-
"size_in_bytes":
|
| 3283 |
},
|
| 3284 |
"principle_A_reconstruction": {
|
| 3285 |
"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",
|
|
|
|
| 3203 |
"features": {
|
| 3204 |
"sentence_good": {
|
| 3205 |
"dtype": "string",
|
|
|
|
| 3206 |
"_type": "Value"
|
| 3207 |
},
|
| 3208 |
"sentence_bad": {
|
| 3209 |
"dtype": "string",
|
|
|
|
| 3210 |
"_type": "Value"
|
| 3211 |
},
|
| 3212 |
"field": {
|
| 3213 |
"dtype": "string",
|
|
|
|
| 3214 |
"_type": "Value"
|
| 3215 |
},
|
| 3216 |
"linguistics_term": {
|
| 3217 |
"dtype": "string",
|
|
|
|
| 3218 |
"_type": "Value"
|
| 3219 |
},
|
| 3220 |
"UID": {
|
| 3221 |
"dtype": "string",
|
|
|
|
| 3222 |
"_type": "Value"
|
| 3223 |
},
|
| 3224 |
"simple_LM_method": {
|
| 3225 |
"dtype": "bool",
|
|
|
|
| 3226 |
"_type": "Value"
|
| 3227 |
},
|
| 3228 |
"one_prefix_method": {
|
| 3229 |
"dtype": "bool",
|
|
|
|
| 3230 |
"_type": "Value"
|
| 3231 |
},
|
| 3232 |
"two_prefix_method": {
|
| 3233 |
"dtype": "bool",
|
|
|
|
| 3234 |
"_type": "Value"
|
| 3235 |
},
|
| 3236 |
"lexically_identical": {
|
| 3237 |
"dtype": "bool",
|
|
|
|
| 3238 |
"_type": "Value"
|
| 3239 |
},
|
| 3240 |
"pair_id": {
|
| 3241 |
"dtype": "int32",
|
|
|
|
| 3242 |
"_type": "Value"
|
| 3243 |
}
|
| 3244 |
},
|
|
|
|
| 3245 |
"builder_name": "blimp",
|
| 3246 |
+
"dataset_name": "blimp",
|
| 3247 |
"config_name": "principle_A_domain_3",
|
| 3248 |
"version": {
|
| 3249 |
"version_str": "0.1.0",
|
|
|
|
|
|
|
| 3250 |
"major": 0,
|
| 3251 |
"minor": 1,
|
| 3252 |
"patch": 0
|
|
|
|
| 3254 |
"splits": {
|
| 3255 |
"train": {
|
| 3256 |
"name": "train",
|
| 3257 |
+
"num_bytes": 158998,
|
| 3258 |
"num_examples": 1000,
|
| 3259 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3260 |
}
|
| 3261 |
},
|
| 3262 |
+
"download_size": 52859,
|
| 3263 |
+
"dataset_size": 158998,
|
| 3264 |
+
"size_in_bytes": 211857
|
| 3265 |
},
|
| 3266 |
"principle_A_reconstruction": {
|
| 3267 |
"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_domain_3/train-00000-of-00001.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:25227d39f6168b88f5b3ee0b35d0f6dd8105545016d76764d11c27f6031844d5
|
| 3 |
+
size 52859
|