Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
·
d458d7f
1
Parent(s):
d7a192b
Add passive_2 data files
Browse files- README.md +7 -3
- dataset_infos.json +6 -24
- passive_2/train-00000-of-00001.parquet +3 -0
README.md
CHANGED
@@ -1192,10 +1192,10 @@ dataset_info:
|
|
1192 |
dtype: int32
|
1193 |
splits:
|
1194 |
- name: train
|
1195 |
-
num_bytes:
|
1196 |
num_examples: 1000
|
1197 |
-
download_size:
|
1198 |
-
dataset_size:
|
1199 |
- config_name: principle_A_c_command
|
1200 |
features:
|
1201 |
- name: sentence_good
|
@@ -2061,6 +2061,10 @@ configs:
|
|
2061 |
data_files:
|
2062 |
- split: train
|
2063 |
path: passive_1/train-*
|
|
|
|
|
|
|
|
|
2064 |
---
|
2065 |
|
2066 |
# Dataset Card for "blimp"
|
|
|
1192 |
dtype: int32
|
1193 |
splits:
|
1194 |
- name: train
|
1195 |
+
num_bytes: 113960
|
1196 |
num_examples: 1000
|
1197 |
+
download_size: 40499
|
1198 |
+
dataset_size: 113960
|
1199 |
- config_name: principle_A_c_command
|
1200 |
features:
|
1201 |
- name: sentence_good
|
|
|
2061 |
data_files:
|
2062 |
- split: train
|
2063 |
path: passive_1/train-*
|
2064 |
+
- config_name: passive_2
|
2065 |
+
data_files:
|
2066 |
+
- split: train
|
2067 |
+
path: passive_2/train-*
|
2068 |
---
|
2069 |
|
2070 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -2795,62 +2795,50 @@
|
|
2795 |
"features": {
|
2796 |
"sentence_good": {
|
2797 |
"dtype": "string",
|
2798 |
-
"id": null,
|
2799 |
"_type": "Value"
|
2800 |
},
|
2801 |
"sentence_bad": {
|
2802 |
"dtype": "string",
|
2803 |
-
"id": null,
|
2804 |
"_type": "Value"
|
2805 |
},
|
2806 |
"field": {
|
2807 |
"dtype": "string",
|
2808 |
-
"id": null,
|
2809 |
"_type": "Value"
|
2810 |
},
|
2811 |
"linguistics_term": {
|
2812 |
"dtype": "string",
|
2813 |
-
"id": null,
|
2814 |
"_type": "Value"
|
2815 |
},
|
2816 |
"UID": {
|
2817 |
"dtype": "string",
|
2818 |
-
"id": null,
|
2819 |
"_type": "Value"
|
2820 |
},
|
2821 |
"simple_LM_method": {
|
2822 |
"dtype": "bool",
|
2823 |
-
"id": null,
|
2824 |
"_type": "Value"
|
2825 |
},
|
2826 |
"one_prefix_method": {
|
2827 |
"dtype": "bool",
|
2828 |
-
"id": null,
|
2829 |
"_type": "Value"
|
2830 |
},
|
2831 |
"two_prefix_method": {
|
2832 |
"dtype": "bool",
|
2833 |
-
"id": null,
|
2834 |
"_type": "Value"
|
2835 |
},
|
2836 |
"lexically_identical": {
|
2837 |
"dtype": "bool",
|
2838 |
-
"id": null,
|
2839 |
"_type": "Value"
|
2840 |
},
|
2841 |
"pair_id": {
|
2842 |
"dtype": "int32",
|
2843 |
-
"id": null,
|
2844 |
"_type": "Value"
|
2845 |
}
|
2846 |
},
|
2847 |
-
"supervised_keys": null,
|
2848 |
"builder_name": "blimp",
|
|
|
2849 |
"config_name": "passive_2",
|
2850 |
"version": {
|
2851 |
"version_str": "0.1.0",
|
2852 |
-
"description": null,
|
2853 |
-
"datasets_version_to_prepare": null,
|
2854 |
"major": 0,
|
2855 |
"minor": 1,
|
2856 |
"patch": 0
|
@@ -2858,20 +2846,14 @@
|
|
2858 |
"splits": {
|
2859 |
"train": {
|
2860 |
"name": "train",
|
2861 |
-
"num_bytes":
|
2862 |
"num_examples": 1000,
|
2863 |
-
"dataset_name":
|
2864 |
-
}
|
2865 |
-
},
|
2866 |
-
"download_checksums": {
|
2867 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/passive_2.jsonl": {
|
2868 |
-
"num_bytes": 308350,
|
2869 |
-
"checksum": "4d49e438185bed4122a252dd9ee69bc8eccb2fe6c1b09ccc78198fd925fed8c5"
|
2870 |
}
|
2871 |
},
|
2872 |
-
"download_size":
|
2873 |
-
"dataset_size":
|
2874 |
-
"size_in_bytes":
|
2875 |
},
|
2876 |
"principle_A_c_command": {
|
2877 |
"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",
|
|
|
2795 |
"features": {
|
2796 |
"sentence_good": {
|
2797 |
"dtype": "string",
|
|
|
2798 |
"_type": "Value"
|
2799 |
},
|
2800 |
"sentence_bad": {
|
2801 |
"dtype": "string",
|
|
|
2802 |
"_type": "Value"
|
2803 |
},
|
2804 |
"field": {
|
2805 |
"dtype": "string",
|
|
|
2806 |
"_type": "Value"
|
2807 |
},
|
2808 |
"linguistics_term": {
|
2809 |
"dtype": "string",
|
|
|
2810 |
"_type": "Value"
|
2811 |
},
|
2812 |
"UID": {
|
2813 |
"dtype": "string",
|
|
|
2814 |
"_type": "Value"
|
2815 |
},
|
2816 |
"simple_LM_method": {
|
2817 |
"dtype": "bool",
|
|
|
2818 |
"_type": "Value"
|
2819 |
},
|
2820 |
"one_prefix_method": {
|
2821 |
"dtype": "bool",
|
|
|
2822 |
"_type": "Value"
|
2823 |
},
|
2824 |
"two_prefix_method": {
|
2825 |
"dtype": "bool",
|
|
|
2826 |
"_type": "Value"
|
2827 |
},
|
2828 |
"lexically_identical": {
|
2829 |
"dtype": "bool",
|
|
|
2830 |
"_type": "Value"
|
2831 |
},
|
2832 |
"pair_id": {
|
2833 |
"dtype": "int32",
|
|
|
2834 |
"_type": "Value"
|
2835 |
}
|
2836 |
},
|
|
|
2837 |
"builder_name": "blimp",
|
2838 |
+
"dataset_name": "blimp",
|
2839 |
"config_name": "passive_2",
|
2840 |
"version": {
|
2841 |
"version_str": "0.1.0",
|
|
|
|
|
2842 |
"major": 0,
|
2843 |
"minor": 1,
|
2844 |
"patch": 0
|
|
|
2846 |
"splits": {
|
2847 |
"train": {
|
2848 |
"name": "train",
|
2849 |
+
"num_bytes": 113960,
|
2850 |
"num_examples": 1000,
|
2851 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
2852 |
}
|
2853 |
},
|
2854 |
+
"download_size": 40499,
|
2855 |
+
"dataset_size": 113960,
|
2856 |
+
"size_in_bytes": 154459
|
2857 |
},
|
2858 |
"principle_A_c_command": {
|
2859 |
"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",
|
passive_2/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ee29e69ca55ec89cf648a57595f13c2b7271995ffb3329e10233a6703dbdd1bf
|
3 |
+
size 40499
|