Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
•
9260d51
1
Parent(s):
80a1d5e
Add only_npi_scope data files
Browse files- README.md +7 -3
- dataset_infos.json +6 -24
- only_npi_scope/train-00000-of-00001.parquet +3 -0
README.md
CHANGED
@@ -1136,10 +1136,10 @@ dataset_info:
|
|
1136 |
dtype: int32
|
1137 |
splits:
|
1138 |
- name: train
|
1139 |
-
num_bytes:
|
1140 |
num_examples: 1000
|
1141 |
-
download_size:
|
1142 |
-
dataset_size:
|
1143 |
- config_name: passive_1
|
1144 |
features:
|
1145 |
- name: sentence_good
|
@@ -2053,6 +2053,10 @@ configs:
|
|
2053 |
data_files:
|
2054 |
- split: train
|
2055 |
path: only_npi_licensor_present/train-*
|
|
|
|
|
|
|
|
|
2056 |
---
|
2057 |
|
2058 |
# Dataset Card for "blimp"
|
|
|
1136 |
dtype: int32
|
1137 |
splits:
|
1138 |
- name: train
|
1139 |
+
num_bytes: 208902
|
1140 |
num_examples: 1000
|
1141 |
+
download_size: 84970
|
1142 |
+
dataset_size: 208902
|
1143 |
- config_name: passive_1
|
1144 |
features:
|
1145 |
- name: sentence_good
|
|
|
2053 |
data_files:
|
2054 |
- split: train
|
2055 |
path: only_npi_licensor_present/train-*
|
2056 |
+
- config_name: only_npi_scope
|
2057 |
+
data_files:
|
2058 |
+
- split: train
|
2059 |
+
path: only_npi_scope/train-*
|
2060 |
---
|
2061 |
|
2062 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -2659,62 +2659,50 @@
|
|
2659 |
"features": {
|
2660 |
"sentence_good": {
|
2661 |
"dtype": "string",
|
2662 |
-
"id": null,
|
2663 |
"_type": "Value"
|
2664 |
},
|
2665 |
"sentence_bad": {
|
2666 |
"dtype": "string",
|
2667 |
-
"id": null,
|
2668 |
"_type": "Value"
|
2669 |
},
|
2670 |
"field": {
|
2671 |
"dtype": "string",
|
2672 |
-
"id": null,
|
2673 |
"_type": "Value"
|
2674 |
},
|
2675 |
"linguistics_term": {
|
2676 |
"dtype": "string",
|
2677 |
-
"id": null,
|
2678 |
"_type": "Value"
|
2679 |
},
|
2680 |
"UID": {
|
2681 |
"dtype": "string",
|
2682 |
-
"id": null,
|
2683 |
"_type": "Value"
|
2684 |
},
|
2685 |
"simple_LM_method": {
|
2686 |
"dtype": "bool",
|
2687 |
-
"id": null,
|
2688 |
"_type": "Value"
|
2689 |
},
|
2690 |
"one_prefix_method": {
|
2691 |
"dtype": "bool",
|
2692 |
-
"id": null,
|
2693 |
"_type": "Value"
|
2694 |
},
|
2695 |
"two_prefix_method": {
|
2696 |
"dtype": "bool",
|
2697 |
-
"id": null,
|
2698 |
"_type": "Value"
|
2699 |
},
|
2700 |
"lexically_identical": {
|
2701 |
"dtype": "bool",
|
2702 |
-
"id": null,
|
2703 |
"_type": "Value"
|
2704 |
},
|
2705 |
"pair_id": {
|
2706 |
"dtype": "int32",
|
2707 |
-
"id": null,
|
2708 |
"_type": "Value"
|
2709 |
}
|
2710 |
},
|
2711 |
-
"supervised_keys": null,
|
2712 |
"builder_name": "blimp",
|
|
|
2713 |
"config_name": "only_npi_scope",
|
2714 |
"version": {
|
2715 |
"version_str": "0.1.0",
|
2716 |
-
"description": null,
|
2717 |
-
"datasets_version_to_prepare": null,
|
2718 |
"major": 0,
|
2719 |
"minor": 1,
|
2720 |
"patch": 0
|
@@ -2722,20 +2710,14 @@
|
|
2722 |
"splits": {
|
2723 |
"train": {
|
2724 |
"name": "train",
|
2725 |
-
"num_bytes":
|
2726 |
"num_examples": 1000,
|
2727 |
-
"dataset_name":
|
2728 |
-
}
|
2729 |
-
},
|
2730 |
-
"download_checksums": {
|
2731 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/only_npi_scope.jsonl": {
|
2732 |
-
"num_bytes": 583720,
|
2733 |
-
"checksum": "ca1f0671643e37936515f02125036bc2904ab9b35c4689b0e7a3ea3914e571f6"
|
2734 |
}
|
2735 |
},
|
2736 |
-
"download_size":
|
2737 |
-
"dataset_size":
|
2738 |
-
"size_in_bytes":
|
2739 |
},
|
2740 |
"passive_1": {
|
2741 |
"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",
|
|
|
2659 |
"features": {
|
2660 |
"sentence_good": {
|
2661 |
"dtype": "string",
|
|
|
2662 |
"_type": "Value"
|
2663 |
},
|
2664 |
"sentence_bad": {
|
2665 |
"dtype": "string",
|
|
|
2666 |
"_type": "Value"
|
2667 |
},
|
2668 |
"field": {
|
2669 |
"dtype": "string",
|
|
|
2670 |
"_type": "Value"
|
2671 |
},
|
2672 |
"linguistics_term": {
|
2673 |
"dtype": "string",
|
|
|
2674 |
"_type": "Value"
|
2675 |
},
|
2676 |
"UID": {
|
2677 |
"dtype": "string",
|
|
|
2678 |
"_type": "Value"
|
2679 |
},
|
2680 |
"simple_LM_method": {
|
2681 |
"dtype": "bool",
|
|
|
2682 |
"_type": "Value"
|
2683 |
},
|
2684 |
"one_prefix_method": {
|
2685 |
"dtype": "bool",
|
|
|
2686 |
"_type": "Value"
|
2687 |
},
|
2688 |
"two_prefix_method": {
|
2689 |
"dtype": "bool",
|
|
|
2690 |
"_type": "Value"
|
2691 |
},
|
2692 |
"lexically_identical": {
|
2693 |
"dtype": "bool",
|
|
|
2694 |
"_type": "Value"
|
2695 |
},
|
2696 |
"pair_id": {
|
2697 |
"dtype": "int32",
|
|
|
2698 |
"_type": "Value"
|
2699 |
}
|
2700 |
},
|
|
|
2701 |
"builder_name": "blimp",
|
2702 |
+
"dataset_name": "blimp",
|
2703 |
"config_name": "only_npi_scope",
|
2704 |
"version": {
|
2705 |
"version_str": "0.1.0",
|
|
|
|
|
2706 |
"major": 0,
|
2707 |
"minor": 1,
|
2708 |
"patch": 0
|
|
|
2710 |
"splits": {
|
2711 |
"train": {
|
2712 |
"name": "train",
|
2713 |
+
"num_bytes": 208902,
|
2714 |
"num_examples": 1000,
|
2715 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
2716 |
}
|
2717 |
},
|
2718 |
+
"download_size": 84970,
|
2719 |
+
"dataset_size": 208902,
|
2720 |
+
"size_in_bytes": 293872
|
2721 |
},
|
2722 |
"passive_1": {
|
2723 |
"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",
|
only_npi_scope/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c5c14f42c274448aaaaa7fe0e94440f9fdbe5c01a9cf1673af824189678d90e7
|
3 |
+
size 84970
|