Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
•
d2b1da7
1
Parent(s):
6d579cb
Add principle_A_case_1 data files
Browse files- README.md +7 -3
- dataset_infos.json +6 -24
- principle_A_case_1/train-00000-of-00001.parquet +3 -0
README.md
CHANGED
@@ -1248,10 +1248,10 @@ dataset_info:
|
|
1248 |
dtype: int32
|
1249 |
splits:
|
1250 |
- name: train
|
1251 |
-
num_bytes:
|
1252 |
num_examples: 1000
|
1253 |
-
download_size:
|
1254 |
-
dataset_size:
|
1255 |
- config_name: principle_A_case_2
|
1256 |
features:
|
1257 |
- name: sentence_good
|
@@ -2069,6 +2069,10 @@ configs:
|
|
2069 |
data_files:
|
2070 |
- split: train
|
2071 |
path: principle_A_c_command/train-*
|
|
|
|
|
|
|
|
|
2072 |
---
|
2073 |
|
2074 |
# Dataset Card for "blimp"
|
|
|
1248 |
dtype: int32
|
1249 |
splits:
|
1250 |
- name: train
|
1251 |
+
num_bytes: 170398
|
1252 |
num_examples: 1000
|
1253 |
+
download_size: 61092
|
1254 |
+
dataset_size: 170398
|
1255 |
- config_name: principle_A_case_2
|
1256 |
features:
|
1257 |
- name: sentence_good
|
|
|
2069 |
data_files:
|
2070 |
- split: train
|
2071 |
path: principle_A_c_command/train-*
|
2072 |
+
- config_name: principle_A_case_1
|
2073 |
+
data_files:
|
2074 |
+
- split: train
|
2075 |
+
path: principle_A_case_1/train-*
|
2076 |
---
|
2077 |
|
2078 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -2931,62 +2931,50 @@
|
|
2931 |
"features": {
|
2932 |
"sentence_good": {
|
2933 |
"dtype": "string",
|
2934 |
-
"id": null,
|
2935 |
"_type": "Value"
|
2936 |
},
|
2937 |
"sentence_bad": {
|
2938 |
"dtype": "string",
|
2939 |
-
"id": null,
|
2940 |
"_type": "Value"
|
2941 |
},
|
2942 |
"field": {
|
2943 |
"dtype": "string",
|
2944 |
-
"id": null,
|
2945 |
"_type": "Value"
|
2946 |
},
|
2947 |
"linguistics_term": {
|
2948 |
"dtype": "string",
|
2949 |
-
"id": null,
|
2950 |
"_type": "Value"
|
2951 |
},
|
2952 |
"UID": {
|
2953 |
"dtype": "string",
|
2954 |
-
"id": null,
|
2955 |
"_type": "Value"
|
2956 |
},
|
2957 |
"simple_LM_method": {
|
2958 |
"dtype": "bool",
|
2959 |
-
"id": null,
|
2960 |
"_type": "Value"
|
2961 |
},
|
2962 |
"one_prefix_method": {
|
2963 |
"dtype": "bool",
|
2964 |
-
"id": null,
|
2965 |
"_type": "Value"
|
2966 |
},
|
2967 |
"two_prefix_method": {
|
2968 |
"dtype": "bool",
|
2969 |
-
"id": null,
|
2970 |
"_type": "Value"
|
2971 |
},
|
2972 |
"lexically_identical": {
|
2973 |
"dtype": "bool",
|
2974 |
-
"id": null,
|
2975 |
"_type": "Value"
|
2976 |
},
|
2977 |
"pair_id": {
|
2978 |
"dtype": "int32",
|
2979 |
-
"id": null,
|
2980 |
"_type": "Value"
|
2981 |
}
|
2982 |
},
|
2983 |
-
"supervised_keys": null,
|
2984 |
"builder_name": "blimp",
|
|
|
2985 |
"config_name": "principle_A_case_1",
|
2986 |
"version": {
|
2987 |
"version_str": "0.1.0",
|
2988 |
-
"description": null,
|
2989 |
-
"datasets_version_to_prepare": null,
|
2990 |
"major": 0,
|
2991 |
"minor": 1,
|
2992 |
"patch": 0
|
@@ -2994,20 +2982,14 @@
|
|
2994 |
"splits": {
|
2995 |
"train": {
|
2996 |
"name": "train",
|
2997 |
-
"num_bytes":
|
2998 |
"num_examples": 1000,
|
2999 |
-
"dataset_name":
|
3000 |
-
}
|
3001 |
-
},
|
3002 |
-
"download_checksums": {
|
3003 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/principle_A_case_1.jsonl": {
|
3004 |
-
"num_bytes": 477239,
|
3005 |
-
"checksum": "7efb5246fe315dc2cf56b80739783cffcddee9e52650716cf5aa0f6987f7644d"
|
3006 |
}
|
3007 |
},
|
3008 |
-
"download_size":
|
3009 |
-
"dataset_size":
|
3010 |
-
"size_in_bytes":
|
3011 |
},
|
3012 |
"principle_A_case_2": {
|
3013 |
"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",
|
|
|
2931 |
"features": {
|
2932 |
"sentence_good": {
|
2933 |
"dtype": "string",
|
|
|
2934 |
"_type": "Value"
|
2935 |
},
|
2936 |
"sentence_bad": {
|
2937 |
"dtype": "string",
|
|
|
2938 |
"_type": "Value"
|
2939 |
},
|
2940 |
"field": {
|
2941 |
"dtype": "string",
|
|
|
2942 |
"_type": "Value"
|
2943 |
},
|
2944 |
"linguistics_term": {
|
2945 |
"dtype": "string",
|
|
|
2946 |
"_type": "Value"
|
2947 |
},
|
2948 |
"UID": {
|
2949 |
"dtype": "string",
|
|
|
2950 |
"_type": "Value"
|
2951 |
},
|
2952 |
"simple_LM_method": {
|
2953 |
"dtype": "bool",
|
|
|
2954 |
"_type": "Value"
|
2955 |
},
|
2956 |
"one_prefix_method": {
|
2957 |
"dtype": "bool",
|
|
|
2958 |
"_type": "Value"
|
2959 |
},
|
2960 |
"two_prefix_method": {
|
2961 |
"dtype": "bool",
|
|
|
2962 |
"_type": "Value"
|
2963 |
},
|
2964 |
"lexically_identical": {
|
2965 |
"dtype": "bool",
|
|
|
2966 |
"_type": "Value"
|
2967 |
},
|
2968 |
"pair_id": {
|
2969 |
"dtype": "int32",
|
|
|
2970 |
"_type": "Value"
|
2971 |
}
|
2972 |
},
|
|
|
2973 |
"builder_name": "blimp",
|
2974 |
+
"dataset_name": "blimp",
|
2975 |
"config_name": "principle_A_case_1",
|
2976 |
"version": {
|
2977 |
"version_str": "0.1.0",
|
|
|
|
|
2978 |
"major": 0,
|
2979 |
"minor": 1,
|
2980 |
"patch": 0
|
|
|
2982 |
"splits": {
|
2983 |
"train": {
|
2984 |
"name": "train",
|
2985 |
+
"num_bytes": 170398,
|
2986 |
"num_examples": 1000,
|
2987 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
2988 |
}
|
2989 |
},
|
2990 |
+
"download_size": 61092,
|
2991 |
+
"dataset_size": 170398,
|
2992 |
+
"size_in_bytes": 231490
|
2993 |
},
|
2994 |
"principle_A_case_2": {
|
2995 |
"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_1/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6615f7f1572b9bcc335be59e7645002ceb2c95d92dc2371551c91db48d9a9a55
|
3 |
+
size 61092
|