Update README.md

README.md

@@ -29,11 +29,15 @@ data_files:
 
 ## Dataset Summary
 
-The agatha_christie dataset contains the complete works of Agatha Christie, including the novels, short stories, and plays. The dataset is in English and contains 66 works in total. The dataset is intended for text generation tasks, such as language modeling, and can be used to train models to generate text in the style of Agatha Christie.
+The agatha_christie dataset contains the complete works of Agatha Christie, 
+including the novels, short stories, and plays. The dataset is in English and contains 
+110 works in total. The dataset is intended for text generation tasks, such as language
+modeling, and can be used to train models to generate text in the style of Agatha Christie.
 
 ## Supported Tasks and Leaderboards
 
-The dataset can be used for text generation tasks, such as language modeling. The dataset can be used to train models to generate text in the style of Agatha Christie.
+The dataset can be used for text generation tasks, such as language modeling. The dataset 
+can be used to train models to generate text in the style of Agatha Christie.
 
 ## Languages
 
@@ -55,4 +59,12 @@ The following is an example sample from the dataset.
 ### Data Fields
 
 - text: The text of the work chunked into semantic segments by llamaindex SemanticNodeParser.
-- source: The source material of the text.
+- source: The source material of the text.
+
+### Splits
+
+The dataset is split into tain, test and validation splits.
+
+- train.parquet
+- test.parquet
+- validation.parquet