Upload dataset

CosmosQA/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a25895793e74478de2ef2dedcd676c9da3dab2354a3ab9a8acec01012b62e99b
+size 7853834

README.md

@@ -45,6 +45,21 @@ dataset_info:
     num_examples: 9427
   download_size: 3550975
   dataset_size: 17836256
+- config_name: CosmosQA
+  features:
+  - name: input_ids
+    sequence: int32
+  - name: input_sentences_ids
+    sequence:
+      sequence: int64
+  - name: labels
+    sequence: int64
+  splits:
+  - name: train
+    num_bytes: 53658116
+    num_examples: 25262
+  download_size: 7853834
+  dataset_size: 53658116
 - config_name: MultiNews
   features:
   - name: input_ids
@@ -193,6 +208,10 @@ configs:
   data_files:
   - split: train
     path: BoolQ/train-*
+- config_name: CosmosQA
+  data_files:
+  - split: train
+    path: CosmosQA/train-*
 - config_name: MultiNews
   data_files:
   - split: train