Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
•
a96b0d3
1
Parent(s):
62b5a85
Add ellipsis_n_bar_2 data files
Browse files- README.md +7 -3
- dataset_infos.json +6 -24
- ellipsis_n_bar_2/train-00000-of-00001.parquet +3 -0
README.md
CHANGED
@@ -632,10 +632,10 @@ dataset_info:
|
|
632 |
dtype: int32
|
633 |
splits:
|
634 |
- name: train
|
635 |
-
num_bytes:
|
636 |
num_examples: 1000
|
637 |
-
download_size:
|
638 |
-
dataset_size:
|
639 |
- config_name: existential_there_object_raising
|
640 |
features:
|
641 |
- name: sentence_good
|
@@ -1981,6 +1981,10 @@ configs:
|
|
1981 |
data_files:
|
1982 |
- split: train
|
1983 |
path: ellipsis_n_bar_1/train-*
|
|
|
|
|
|
|
|
|
1984 |
---
|
1985 |
|
1986 |
# Dataset Card for "blimp"
|
|
|
632 |
dtype: int32
|
633 |
splits:
|
634 |
- name: train
|
635 |
+
num_bytes: 233161
|
636 |
num_examples: 1000
|
637 |
+
download_size: 98882
|
638 |
+
dataset_size: 233161
|
639 |
- config_name: existential_there_object_raising
|
640 |
features:
|
641 |
- name: sentence_good
|
|
|
1981 |
data_files:
|
1982 |
- split: train
|
1983 |
path: ellipsis_n_bar_1/train-*
|
1984 |
+
- config_name: ellipsis_n_bar_2
|
1985 |
+
data_files:
|
1986 |
+
- split: train
|
1987 |
+
path: ellipsis_n_bar_2/train-*
|
1988 |
---
|
1989 |
|
1990 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -1435,62 +1435,50 @@
|
|
1435 |
"features": {
|
1436 |
"sentence_good": {
|
1437 |
"dtype": "string",
|
1438 |
-
"id": null,
|
1439 |
"_type": "Value"
|
1440 |
},
|
1441 |
"sentence_bad": {
|
1442 |
"dtype": "string",
|
1443 |
-
"id": null,
|
1444 |
"_type": "Value"
|
1445 |
},
|
1446 |
"field": {
|
1447 |
"dtype": "string",
|
1448 |
-
"id": null,
|
1449 |
"_type": "Value"
|
1450 |
},
|
1451 |
"linguistics_term": {
|
1452 |
"dtype": "string",
|
1453 |
-
"id": null,
|
1454 |
"_type": "Value"
|
1455 |
},
|
1456 |
"UID": {
|
1457 |
"dtype": "string",
|
1458 |
-
"id": null,
|
1459 |
"_type": "Value"
|
1460 |
},
|
1461 |
"simple_LM_method": {
|
1462 |
"dtype": "bool",
|
1463 |
-
"id": null,
|
1464 |
"_type": "Value"
|
1465 |
},
|
1466 |
"one_prefix_method": {
|
1467 |
"dtype": "bool",
|
1468 |
-
"id": null,
|
1469 |
"_type": "Value"
|
1470 |
},
|
1471 |
"two_prefix_method": {
|
1472 |
"dtype": "bool",
|
1473 |
-
"id": null,
|
1474 |
"_type": "Value"
|
1475 |
},
|
1476 |
"lexically_identical": {
|
1477 |
"dtype": "bool",
|
1478 |
-
"id": null,
|
1479 |
"_type": "Value"
|
1480 |
},
|
1481 |
"pair_id": {
|
1482 |
"dtype": "int32",
|
1483 |
-
"id": null,
|
1484 |
"_type": "Value"
|
1485 |
}
|
1486 |
},
|
1487 |
-
"supervised_keys": null,
|
1488 |
"builder_name": "blimp",
|
|
|
1489 |
"config_name": "ellipsis_n_bar_2",
|
1490 |
"version": {
|
1491 |
"version_str": "0.1.0",
|
1492 |
-
"description": null,
|
1493 |
-
"datasets_version_to_prepare": null,
|
1494 |
"major": 0,
|
1495 |
"minor": 1,
|
1496 |
"patch": 0
|
@@ -1498,20 +1486,14 @@
|
|
1498 |
"splits": {
|
1499 |
"train": {
|
1500 |
"name": "train",
|
1501 |
-
"num_bytes":
|
1502 |
"num_examples": 1000,
|
1503 |
-
"dataset_name":
|
1504 |
-
}
|
1505 |
-
},
|
1506 |
-
"download_checksums": {
|
1507 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/ellipsis_n_bar_2.jsonl": {
|
1508 |
-
"num_bytes": 427551,
|
1509 |
-
"checksum": "3dfc2b378ea4867186917ec9cce5ce927e4d3bb6addf018bc31b870dd2aa8474"
|
1510 |
}
|
1511 |
},
|
1512 |
-
"download_size":
|
1513 |
-
"dataset_size":
|
1514 |
-
"size_in_bytes":
|
1515 |
},
|
1516 |
"existential_there_object_raising": {
|
1517 |
"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",
|
|
|
1435 |
"features": {
|
1436 |
"sentence_good": {
|
1437 |
"dtype": "string",
|
|
|
1438 |
"_type": "Value"
|
1439 |
},
|
1440 |
"sentence_bad": {
|
1441 |
"dtype": "string",
|
|
|
1442 |
"_type": "Value"
|
1443 |
},
|
1444 |
"field": {
|
1445 |
"dtype": "string",
|
|
|
1446 |
"_type": "Value"
|
1447 |
},
|
1448 |
"linguistics_term": {
|
1449 |
"dtype": "string",
|
|
|
1450 |
"_type": "Value"
|
1451 |
},
|
1452 |
"UID": {
|
1453 |
"dtype": "string",
|
|
|
1454 |
"_type": "Value"
|
1455 |
},
|
1456 |
"simple_LM_method": {
|
1457 |
"dtype": "bool",
|
|
|
1458 |
"_type": "Value"
|
1459 |
},
|
1460 |
"one_prefix_method": {
|
1461 |
"dtype": "bool",
|
|
|
1462 |
"_type": "Value"
|
1463 |
},
|
1464 |
"two_prefix_method": {
|
1465 |
"dtype": "bool",
|
|
|
1466 |
"_type": "Value"
|
1467 |
},
|
1468 |
"lexically_identical": {
|
1469 |
"dtype": "bool",
|
|
|
1470 |
"_type": "Value"
|
1471 |
},
|
1472 |
"pair_id": {
|
1473 |
"dtype": "int32",
|
|
|
1474 |
"_type": "Value"
|
1475 |
}
|
1476 |
},
|
|
|
1477 |
"builder_name": "blimp",
|
1478 |
+
"dataset_name": "blimp",
|
1479 |
"config_name": "ellipsis_n_bar_2",
|
1480 |
"version": {
|
1481 |
"version_str": "0.1.0",
|
|
|
|
|
1482 |
"major": 0,
|
1483 |
"minor": 1,
|
1484 |
"patch": 0
|
|
|
1486 |
"splits": {
|
1487 |
"train": {
|
1488 |
"name": "train",
|
1489 |
+
"num_bytes": 233161,
|
1490 |
"num_examples": 1000,
|
1491 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
1492 |
}
|
1493 |
},
|
1494 |
+
"download_size": 98882,
|
1495 |
+
"dataset_size": 233161,
|
1496 |
+
"size_in_bytes": 332043
|
1497 |
},
|
1498 |
"existential_there_object_raising": {
|
1499 |
"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",
|
ellipsis_n_bar_2/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7506bd80abb5c2d54afddf64042bc2df6d587b271f37832c2dcdfa1b66c11d70
|
3 |
+
size 98882
|