Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
•
c469a80
1
Parent(s):
d2b1da7
Add principle_A_case_2 data files
Browse files- README.md +7 -3
- dataset_infos.json +6 -24
- principle_A_case_2/train-00000-of-00001.parquet +3 -0
README.md
CHANGED
@@ -1276,10 +1276,10 @@ dataset_info:
|
|
1276 |
dtype: int32
|
1277 |
splits:
|
1278 |
- name: train
|
1279 |
-
num_bytes:
|
1280 |
num_examples: 1000
|
1281 |
-
download_size:
|
1282 |
-
dataset_size:
|
1283 |
- config_name: principle_A_domain_1
|
1284 |
features:
|
1285 |
- name: sentence_good
|
@@ -2073,6 +2073,10 @@ configs:
|
|
2073 |
data_files:
|
2074 |
- split: train
|
2075 |
path: principle_A_case_1/train-*
|
|
|
|
|
|
|
|
|
2076 |
---
|
2077 |
|
2078 |
# Dataset Card for "blimp"
|
|
|
1276 |
dtype: int32
|
1277 |
splits:
|
1278 |
- name: train
|
1279 |
+
num_bytes: 170412
|
1280 |
num_examples: 1000
|
1281 |
+
download_size: 56430
|
1282 |
+
dataset_size: 170412
|
1283 |
- config_name: principle_A_domain_1
|
1284 |
features:
|
1285 |
- name: sentence_good
|
|
|
2073 |
data_files:
|
2074 |
- split: train
|
2075 |
path: principle_A_case_1/train-*
|
2076 |
+
- config_name: principle_A_case_2
|
2077 |
+
data_files:
|
2078 |
+
- split: train
|
2079 |
+
path: principle_A_case_2/train-*
|
2080 |
---
|
2081 |
|
2082 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -2999,62 +2999,50 @@
|
|
2999 |
"features": {
|
3000 |
"sentence_good": {
|
3001 |
"dtype": "string",
|
3002 |
-
"id": null,
|
3003 |
"_type": "Value"
|
3004 |
},
|
3005 |
"sentence_bad": {
|
3006 |
"dtype": "string",
|
3007 |
-
"id": null,
|
3008 |
"_type": "Value"
|
3009 |
},
|
3010 |
"field": {
|
3011 |
"dtype": "string",
|
3012 |
-
"id": null,
|
3013 |
"_type": "Value"
|
3014 |
},
|
3015 |
"linguistics_term": {
|
3016 |
"dtype": "string",
|
3017 |
-
"id": null,
|
3018 |
"_type": "Value"
|
3019 |
},
|
3020 |
"UID": {
|
3021 |
"dtype": "string",
|
3022 |
-
"id": null,
|
3023 |
"_type": "Value"
|
3024 |
},
|
3025 |
"simple_LM_method": {
|
3026 |
"dtype": "bool",
|
3027 |
-
"id": null,
|
3028 |
"_type": "Value"
|
3029 |
},
|
3030 |
"one_prefix_method": {
|
3031 |
"dtype": "bool",
|
3032 |
-
"id": null,
|
3033 |
"_type": "Value"
|
3034 |
},
|
3035 |
"two_prefix_method": {
|
3036 |
"dtype": "bool",
|
3037 |
-
"id": null,
|
3038 |
"_type": "Value"
|
3039 |
},
|
3040 |
"lexically_identical": {
|
3041 |
"dtype": "bool",
|
3042 |
-
"id": null,
|
3043 |
"_type": "Value"
|
3044 |
},
|
3045 |
"pair_id": {
|
3046 |
"dtype": "int32",
|
3047 |
-
"id": null,
|
3048 |
"_type": "Value"
|
3049 |
}
|
3050 |
},
|
3051 |
-
"supervised_keys": null,
|
3052 |
"builder_name": "blimp",
|
|
|
3053 |
"config_name": "principle_A_case_2",
|
3054 |
"version": {
|
3055 |
"version_str": "0.1.0",
|
3056 |
-
"description": null,
|
3057 |
-
"datasets_version_to_prepare": null,
|
3058 |
"major": 0,
|
3059 |
"minor": 1,
|
3060 |
"patch": 0
|
@@ -3062,20 +3050,14 @@
|
|
3062 |
"splits": {
|
3063 |
"train": {
|
3064 |
"name": "train",
|
3065 |
-
"num_bytes":
|
3066 |
"num_examples": 1000,
|
3067 |
-
"dataset_name":
|
3068 |
-
}
|
3069 |
-
},
|
3070 |
-
"download_checksums": {
|
3071 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/principle_A_case_2.jsonl": {
|
3072 |
-
"num_bytes": 492973,
|
3073 |
-
"checksum": "5de6db4c21a8433bc1f1030e81d5ee151efce02078460f7de7ca5e9566afd914"
|
3074 |
}
|
3075 |
},
|
3076 |
-
"download_size":
|
3077 |
-
"dataset_size":
|
3078 |
-
"size_in_bytes":
|
3079 |
},
|
3080 |
"principle_A_domain_1": {
|
3081 |
"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",
|
|
|
2999 |
"features": {
|
3000 |
"sentence_good": {
|
3001 |
"dtype": "string",
|
|
|
3002 |
"_type": "Value"
|
3003 |
},
|
3004 |
"sentence_bad": {
|
3005 |
"dtype": "string",
|
|
|
3006 |
"_type": "Value"
|
3007 |
},
|
3008 |
"field": {
|
3009 |
"dtype": "string",
|
|
|
3010 |
"_type": "Value"
|
3011 |
},
|
3012 |
"linguistics_term": {
|
3013 |
"dtype": "string",
|
|
|
3014 |
"_type": "Value"
|
3015 |
},
|
3016 |
"UID": {
|
3017 |
"dtype": "string",
|
|
|
3018 |
"_type": "Value"
|
3019 |
},
|
3020 |
"simple_LM_method": {
|
3021 |
"dtype": "bool",
|
|
|
3022 |
"_type": "Value"
|
3023 |
},
|
3024 |
"one_prefix_method": {
|
3025 |
"dtype": "bool",
|
|
|
3026 |
"_type": "Value"
|
3027 |
},
|
3028 |
"two_prefix_method": {
|
3029 |
"dtype": "bool",
|
|
|
3030 |
"_type": "Value"
|
3031 |
},
|
3032 |
"lexically_identical": {
|
3033 |
"dtype": "bool",
|
|
|
3034 |
"_type": "Value"
|
3035 |
},
|
3036 |
"pair_id": {
|
3037 |
"dtype": "int32",
|
|
|
3038 |
"_type": "Value"
|
3039 |
}
|
3040 |
},
|
|
|
3041 |
"builder_name": "blimp",
|
3042 |
+
"dataset_name": "blimp",
|
3043 |
"config_name": "principle_A_case_2",
|
3044 |
"version": {
|
3045 |
"version_str": "0.1.0",
|
|
|
|
|
3046 |
"major": 0,
|
3047 |
"minor": 1,
|
3048 |
"patch": 0
|
|
|
3050 |
"splits": {
|
3051 |
"train": {
|
3052 |
"name": "train",
|
3053 |
+
"num_bytes": 170412,
|
3054 |
"num_examples": 1000,
|
3055 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
3056 |
}
|
3057 |
},
|
3058 |
+
"download_size": 56430,
|
3059 |
+
"dataset_size": 170412,
|
3060 |
+
"size_in_bytes": 226842
|
3061 |
},
|
3062 |
"principle_A_domain_1": {
|
3063 |
"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_case_2/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e3a13d2ab24f231adbc9b3dda2fcd448883855199419db178a449c42eee1d097
|
3 |
+
size 56430
|