Update README.md

README.md

@@ -17,7 +17,7 @@ with no humans labelling them in any way (which is why it can use lots of public
 process to generate inputs and labels from those texts using the BERT base model. 
 
 This is a classification model that solves Short Question Answer Assessment task, finetuned [pretrained DistilBERT model](https://huggingface.co/distilbert-base-uncased) on 
-[Fake and real news dataset](#)
+[Question Answer Assessment dataset](#)
 
 ## Intended uses & limitations
 
@@ -119,7 +119,7 @@ Here is the scores during the training:
 
 ## Evaluation results
 
-When fine-tuned on downstream task of fake news binary classification, this model achieved the following results:
+When fine-tuned on downstream task of Question Answer Assessment, 4 class classification, this model achieved the following results:
 (scores are rounded to 2 floating points)