Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
•
b665065
1
Parent(s):
ff8e6b7
Add matrix_question_npi_licensor_present data files
Browse files
README.md
CHANGED
@@ -1024,10 +1024,10 @@ dataset_info:
|
|
1024 |
dtype: int32
|
1025 |
splits:
|
1026 |
- name: train
|
1027 |
-
num_bytes:
|
1028 |
num_examples: 1000
|
1029 |
-
download_size:
|
1030 |
-
dataset_size:
|
1031 |
- config_name: npi_present_1
|
1032 |
features:
|
1033 |
- name: sentence_good
|
@@ -2037,6 +2037,10 @@ configs:
|
|
2037 |
data_files:
|
2038 |
- split: train
|
2039 |
path: left_branch_island_simple_question/train-*
|
|
|
|
|
|
|
|
|
2040 |
---
|
2041 |
|
2042 |
# Dataset Card for "blimp"
|
|
|
1024 |
dtype: int32
|
1025 |
splits:
|
1026 |
- name: train
|
1027 |
+
num_bytes: 153262
|
1028 |
num_examples: 1000
|
1029 |
+
download_size: 51899
|
1030 |
+
dataset_size: 153262
|
1031 |
- config_name: npi_present_1
|
1032 |
features:
|
1033 |
- name: sentence_good
|
|
|
2037 |
data_files:
|
2038 |
- split: train
|
2039 |
path: left_branch_island_simple_question/train-*
|
2040 |
+
- config_name: matrix_question_npi_licensor_present
|
2041 |
+
data_files:
|
2042 |
+
- split: train
|
2043 |
+
path: matrix_question_npi_licensor_present/train-*
|
2044 |
---
|
2045 |
|
2046 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -2387,62 +2387,50 @@
|
|
2387 |
"features": {
|
2388 |
"sentence_good": {
|
2389 |
"dtype": "string",
|
2390 |
-
"id": null,
|
2391 |
"_type": "Value"
|
2392 |
},
|
2393 |
"sentence_bad": {
|
2394 |
"dtype": "string",
|
2395 |
-
"id": null,
|
2396 |
"_type": "Value"
|
2397 |
},
|
2398 |
"field": {
|
2399 |
"dtype": "string",
|
2400 |
-
"id": null,
|
2401 |
"_type": "Value"
|
2402 |
},
|
2403 |
"linguistics_term": {
|
2404 |
"dtype": "string",
|
2405 |
-
"id": null,
|
2406 |
"_type": "Value"
|
2407 |
},
|
2408 |
"UID": {
|
2409 |
"dtype": "string",
|
2410 |
-
"id": null,
|
2411 |
"_type": "Value"
|
2412 |
},
|
2413 |
"simple_LM_method": {
|
2414 |
"dtype": "bool",
|
2415 |
-
"id": null,
|
2416 |
"_type": "Value"
|
2417 |
},
|
2418 |
"one_prefix_method": {
|
2419 |
"dtype": "bool",
|
2420 |
-
"id": null,
|
2421 |
"_type": "Value"
|
2422 |
},
|
2423 |
"two_prefix_method": {
|
2424 |
"dtype": "bool",
|
2425 |
-
"id": null,
|
2426 |
"_type": "Value"
|
2427 |
},
|
2428 |
"lexically_identical": {
|
2429 |
"dtype": "bool",
|
2430 |
-
"id": null,
|
2431 |
"_type": "Value"
|
2432 |
},
|
2433 |
"pair_id": {
|
2434 |
"dtype": "int32",
|
2435 |
-
"id": null,
|
2436 |
"_type": "Value"
|
2437 |
}
|
2438 |
},
|
2439 |
-
"supervised_keys": null,
|
2440 |
"builder_name": "blimp",
|
|
|
2441 |
"config_name": "matrix_question_npi_licensor_present",
|
2442 |
"version": {
|
2443 |
"version_str": "0.1.0",
|
2444 |
-
"description": null,
|
2445 |
-
"datasets_version_to_prepare": null,
|
2446 |
"major": 0,
|
2447 |
"minor": 1,
|
2448 |
"patch": 0
|
@@ -2450,20 +2438,14 @@
|
|
2450 |
"splits": {
|
2451 |
"train": {
|
2452 |
"name": "train",
|
2453 |
-
"num_bytes":
|
2454 |
"num_examples": 1000,
|
2455 |
-
"dataset_name":
|
2456 |
-
}
|
2457 |
-
},
|
2458 |
-
"download_checksums": {
|
2459 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/matrix_question_npi_licensor_present.jsonl": {
|
2460 |
-
"num_bytes": 457806,
|
2461 |
-
"checksum": "657c9d6e6d50a9cd651881fcbc57cef2fd998611f6f53158ed05e48822fcd2b5"
|
2462 |
}
|
2463 |
},
|
2464 |
-
"download_size":
|
2465 |
-
"dataset_size":
|
2466 |
-
"size_in_bytes":
|
2467 |
},
|
2468 |
"npi_present_1": {
|
2469 |
"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",
|
|
|
2387 |
"features": {
|
2388 |
"sentence_good": {
|
2389 |
"dtype": "string",
|
|
|
2390 |
"_type": "Value"
|
2391 |
},
|
2392 |
"sentence_bad": {
|
2393 |
"dtype": "string",
|
|
|
2394 |
"_type": "Value"
|
2395 |
},
|
2396 |
"field": {
|
2397 |
"dtype": "string",
|
|
|
2398 |
"_type": "Value"
|
2399 |
},
|
2400 |
"linguistics_term": {
|
2401 |
"dtype": "string",
|
|
|
2402 |
"_type": "Value"
|
2403 |
},
|
2404 |
"UID": {
|
2405 |
"dtype": "string",
|
|
|
2406 |
"_type": "Value"
|
2407 |
},
|
2408 |
"simple_LM_method": {
|
2409 |
"dtype": "bool",
|
|
|
2410 |
"_type": "Value"
|
2411 |
},
|
2412 |
"one_prefix_method": {
|
2413 |
"dtype": "bool",
|
|
|
2414 |
"_type": "Value"
|
2415 |
},
|
2416 |
"two_prefix_method": {
|
2417 |
"dtype": "bool",
|
|
|
2418 |
"_type": "Value"
|
2419 |
},
|
2420 |
"lexically_identical": {
|
2421 |
"dtype": "bool",
|
|
|
2422 |
"_type": "Value"
|
2423 |
},
|
2424 |
"pair_id": {
|
2425 |
"dtype": "int32",
|
|
|
2426 |
"_type": "Value"
|
2427 |
}
|
2428 |
},
|
|
|
2429 |
"builder_name": "blimp",
|
2430 |
+
"dataset_name": "blimp",
|
2431 |
"config_name": "matrix_question_npi_licensor_present",
|
2432 |
"version": {
|
2433 |
"version_str": "0.1.0",
|
|
|
|
|
2434 |
"major": 0,
|
2435 |
"minor": 1,
|
2436 |
"patch": 0
|
|
|
2438 |
"splits": {
|
2439 |
"train": {
|
2440 |
"name": "train",
|
2441 |
+
"num_bytes": 153262,
|
2442 |
"num_examples": 1000,
|
2443 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
2444 |
}
|
2445 |
},
|
2446 |
+
"download_size": 51899,
|
2447 |
+
"dataset_size": 153262,
|
2448 |
+
"size_in_bytes": 205161
|
2449 |
},
|
2450 |
"npi_present_1": {
|
2451 |
"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",
|
matrix_question_npi_licensor_present/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a7ed0834104a8fe81b389072c856c10779d660295f924067362740d7901d3532
|
3 |
+
size 51899
|