Merge ^shard_0[1-2]$ into subset_400K

README.md

@@ -1,7 +1,7 @@
 ---
 dataset_info:
 - config_name: shard_01
-  features:
+  features: &id001
   - name: text
     dtype: string
   - name: key
@@ -144,6 +144,11 @@ dataset_info:
     num_examples: 127005
   download_size: 80461130
   dataset_size: 133430843
+- config_name: subset_400K
+  features: *id001
+  splits:
+  - name: train
+    num_examples: 400000
 configs:
 - config_name: shard_01
   data_files:
@@ -193,4 +198,10 @@ configs:
   data_files:
   - split: train
     path: shard_12/train-*
+- config_name: subset_400K
+  data_files:
+  - split: train
+    path:
+    - shard_01/train-*
+    - shard_02/train-*
 ---