Update README.md

Revised explanation under Dataset structure

README.md

@@ -38,9 +38,15 @@ The data is intended to be used for testing and finetuning models for Norwegian
 ## Dataset Structure
 
 ### Data Instances
-Each data instance contains the following features: correct, wrong and category. An instance from the dataset looks like the following:
+Each data instance contains the following features: correct, wrong and category. An instance from the dataset that require a comma looks like the following:
 ```
 {"correct": "Eva klatrer, og Ola går på ski.", "wrong": "Eva klatrer og Ola går på ski.", "category": 1}
+```
+
+An instance with a sentence that is correct without a comma looks like the following:
+
+```
+{"correct": "Husleien går opp og har nå doblet seg på ti år.", "wrong": "Husleien går opp, og har nå doblet seg på ti år.", "category": 1}
 
 ```
 
@@ -51,8 +57,7 @@ Each data instance contains the following features: correct, wrong and category.
 - 'Category': an integer indicating the rule tested by the pair of correct and wrong sentences.
 
 ### Distribution
-The dataset consists of 840 sentences, 600 of which require commas to be grammatically correct, and 240 are correct without any commas. Of the 600 sentences that require commas, most need only one, while 207 require two commas.
-Five sentences utilize a comma as a decimal separator in addition to the grammatical commas.
+The dataset includes 840 instances, each consisting of a pair of sentences: one correctly punctuated and the other incorrectly punctuated. Of these, 600 instances feature sentences that are grammatically correct with the inclusion of commas, while 240 instances contain sentences that are correct without any commas. Among the 600 sentences that require commas, the majority need just one comma, whereas 207 instances require the inclusion of two commas. Five sentences utilize a comma as a decimal separator in addition to the grammatical commas.
 
 
 ### Norwegian comma rules