corrected grammar

README.md

@@ -35,7 +35,7 @@ Repository: **[Github code for SFT Fine-tuning on MathDial](https://github.com/e
 
 Training input and output:
 The model was fine-tuned on the **[MathDial dataset](https://huggingface.co/datasets/eth-nlped/mathdial-chat/viewer/default/train?views%5B%5D=train&row=0)**. 
-Each training example consisted of a **Instruction**, **Student's Name**, **Math Word Problem and Solution**, **The students initial approach**  as input and the **tutor’s step-by-step solution** as the target output.  
+Each training example consisted of a **Instruction**, **Student's Name**, **Math Word Problem and Solution** and **The students initial approach** as input, followed by the **tutor’s step-by-step solution** as the target output.  
 To incorporate the whole conversation, a sliding window approach was used. Every input has the same format:  
 For each step in a conversation, the model input included **all previous turns** in the dialogue (sliding window), followed by the student’s next message. The model’s output was then the **next tutor response** from the dataset. 
 This approach ensures the model learns to generate responses that are context-aware.