Update README.md

README.md

@@ -2,23 +2,16 @@
 configs:
 - config_name: bytelevel
   default: true
-  data_files: "bytelevel/*.parquet"
-  splits:
-  - name: train
+  data_files:
+  - split: train
+    path: "bytelevel/*.parquet"
 - config_name: bytelevel-subset
-  features:
-  - name: id
-    dtype: string
-  - name: input_ids
-    sequence: int32
-  - name: num_tokens
-    dtype: int64
-  splits:
-  - name: train
-    num_bytes: 1924186924
-    num_examples: 100000
-  download_size: 295007200
-  dataset_size: 1924186924
+  default: true
+  data_files:
+  - split: train
+    path: "bytelevel-subset/*.parquet"
+  - split: ngram
+    path: "bytelevel-subset-ngram/*.parquet"
 language:
 - en
 tags: