Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
·
2034afe
1
Parent(s):
6242f11
Add sentential_negation_npi_scope data files
Browse files
README.md
CHANGED
@@ -1500,10 +1500,10 @@ dataset_info:
|
|
1500 |
dtype: int32
|
1501 |
splits:
|
1502 |
- name: train
|
1503 |
-
num_bytes:
|
1504 |
num_examples: 1000
|
1505 |
-
download_size:
|
1506 |
-
dataset_size:
|
1507 |
- config_name: sentential_subject_island
|
1508 |
features:
|
1509 |
- name: sentence_good
|
@@ -2105,6 +2105,10 @@ configs:
|
|
2105 |
data_files:
|
2106 |
- split: train
|
2107 |
path: sentential_negation_npi_licensor_present/train-*
|
|
|
|
|
|
|
|
|
2108 |
---
|
2109 |
|
2110 |
# Dataset Card for "blimp"
|
|
|
1500 |
dtype: int32
|
1501 |
splits:
|
1502 |
- name: train
|
1503 |
+
num_bytes: 232098
|
1504 |
num_examples: 1000
|
1505 |
+
download_size: 90157
|
1506 |
+
dataset_size: 232098
|
1507 |
- config_name: sentential_subject_island
|
1508 |
features:
|
1509 |
- name: sentence_good
|
|
|
2105 |
data_files:
|
2106 |
- split: train
|
2107 |
path: sentential_negation_npi_licensor_present/train-*
|
2108 |
+
- config_name: sentential_negation_npi_scope
|
2109 |
+
data_files:
|
2110 |
+
- split: train
|
2111 |
+
path: sentential_negation_npi_scope/train-*
|
2112 |
---
|
2113 |
|
2114 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -3543,62 +3543,50 @@
|
|
3543 |
"features": {
|
3544 |
"sentence_good": {
|
3545 |
"dtype": "string",
|
3546 |
-
"id": null,
|
3547 |
"_type": "Value"
|
3548 |
},
|
3549 |
"sentence_bad": {
|
3550 |
"dtype": "string",
|
3551 |
-
"id": null,
|
3552 |
"_type": "Value"
|
3553 |
},
|
3554 |
"field": {
|
3555 |
"dtype": "string",
|
3556 |
-
"id": null,
|
3557 |
"_type": "Value"
|
3558 |
},
|
3559 |
"linguistics_term": {
|
3560 |
"dtype": "string",
|
3561 |
-
"id": null,
|
3562 |
"_type": "Value"
|
3563 |
},
|
3564 |
"UID": {
|
3565 |
"dtype": "string",
|
3566 |
-
"id": null,
|
3567 |
"_type": "Value"
|
3568 |
},
|
3569 |
"simple_LM_method": {
|
3570 |
"dtype": "bool",
|
3571 |
-
"id": null,
|
3572 |
"_type": "Value"
|
3573 |
},
|
3574 |
"one_prefix_method": {
|
3575 |
"dtype": "bool",
|
3576 |
-
"id": null,
|
3577 |
"_type": "Value"
|
3578 |
},
|
3579 |
"two_prefix_method": {
|
3580 |
"dtype": "bool",
|
3581 |
-
"id": null,
|
3582 |
"_type": "Value"
|
3583 |
},
|
3584 |
"lexically_identical": {
|
3585 |
"dtype": "bool",
|
3586 |
-
"id": null,
|
3587 |
"_type": "Value"
|
3588 |
},
|
3589 |
"pair_id": {
|
3590 |
"dtype": "int32",
|
3591 |
-
"id": null,
|
3592 |
"_type": "Value"
|
3593 |
}
|
3594 |
},
|
3595 |
-
"supervised_keys": null,
|
3596 |
"builder_name": "blimp",
|
|
|
3597 |
"config_name": "sentential_negation_npi_scope",
|
3598 |
"version": {
|
3599 |
"version_str": "0.1.0",
|
3600 |
-
"description": null,
|
3601 |
-
"datasets_version_to_prepare": null,
|
3602 |
"major": 0,
|
3603 |
"minor": 1,
|
3604 |
"patch": 0
|
@@ -3606,20 +3594,14 @@
|
|
3606 |
"splits": {
|
3607 |
"train": {
|
3608 |
"name": "train",
|
3609 |
-
"num_bytes":
|
3610 |
"num_examples": 1000,
|
3611 |
-
"dataset_name":
|
3612 |
-
}
|
3613 |
-
},
|
3614 |
-
"download_checksums": {
|
3615 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/sentential_negation_npi_scope.jsonl": {
|
3616 |
-
"num_bytes": 614930,
|
3617 |
-
"checksum": "6c341ccc39c94bea45c0412986751bf2f6eca48626728e30e23d2b6a22c8b949"
|
3618 |
}
|
3619 |
},
|
3620 |
-
"download_size":
|
3621 |
-
"dataset_size":
|
3622 |
-
"size_in_bytes":
|
3623 |
},
|
3624 |
"sentential_subject_island": {
|
3625 |
"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",
|
|
|
3543 |
"features": {
|
3544 |
"sentence_good": {
|
3545 |
"dtype": "string",
|
|
|
3546 |
"_type": "Value"
|
3547 |
},
|
3548 |
"sentence_bad": {
|
3549 |
"dtype": "string",
|
|
|
3550 |
"_type": "Value"
|
3551 |
},
|
3552 |
"field": {
|
3553 |
"dtype": "string",
|
|
|
3554 |
"_type": "Value"
|
3555 |
},
|
3556 |
"linguistics_term": {
|
3557 |
"dtype": "string",
|
|
|
3558 |
"_type": "Value"
|
3559 |
},
|
3560 |
"UID": {
|
3561 |
"dtype": "string",
|
|
|
3562 |
"_type": "Value"
|
3563 |
},
|
3564 |
"simple_LM_method": {
|
3565 |
"dtype": "bool",
|
|
|
3566 |
"_type": "Value"
|
3567 |
},
|
3568 |
"one_prefix_method": {
|
3569 |
"dtype": "bool",
|
|
|
3570 |
"_type": "Value"
|
3571 |
},
|
3572 |
"two_prefix_method": {
|
3573 |
"dtype": "bool",
|
|
|
3574 |
"_type": "Value"
|
3575 |
},
|
3576 |
"lexically_identical": {
|
3577 |
"dtype": "bool",
|
|
|
3578 |
"_type": "Value"
|
3579 |
},
|
3580 |
"pair_id": {
|
3581 |
"dtype": "int32",
|
|
|
3582 |
"_type": "Value"
|
3583 |
}
|
3584 |
},
|
|
|
3585 |
"builder_name": "blimp",
|
3586 |
+
"dataset_name": "blimp",
|
3587 |
"config_name": "sentential_negation_npi_scope",
|
3588 |
"version": {
|
3589 |
"version_str": "0.1.0",
|
|
|
|
|
3590 |
"major": 0,
|
3591 |
"minor": 1,
|
3592 |
"patch": 0
|
|
|
3594 |
"splits": {
|
3595 |
"train": {
|
3596 |
"name": "train",
|
3597 |
+
"num_bytes": 232098,
|
3598 |
"num_examples": 1000,
|
3599 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
3600 |
}
|
3601 |
},
|
3602 |
+
"download_size": 90157,
|
3603 |
+
"dataset_size": 232098,
|
3604 |
+
"size_in_bytes": 322255
|
3605 |
},
|
3606 |
"sentential_subject_island": {
|
3607 |
"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",
|
sentential_negation_npi_scope/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:06afb479723e7bff4071ca2f1511c1a78b8849331d09c608aa963fee057f8ac2
|
3 |
+
size 90157
|