Update README.md

README.md

@@ -5,12 +5,17 @@ license: mit
 
 We provide the train, dev, and test sets. For more details, find our report [here](https://github.com/rish-16/cs4248-project/blob/main/CS4248_Group19_Final_Report.pdf). 
 
+## Dataset details
+
+
+
+## Download Instructions
 To access MLe-SNLI, you can use the HuggingFace Datasets API to load the dataset:
 
 ```python
 from datasets import load_dataset
 
-mle_snli = load_dataset("rish-16/MLe-SNLI") # loads a DatasetDict object
+mle_snli = load_dataset("rish16/MLe-SNLI") # loads a DatasetDict object
 train_data = mle_snli['train'] # 500K samples (100K per lang)
 dev_data = mle_snli['dev'] # 49120 samples (9824 per lang)
 test_data = mle_snli['test'] # 49120 samples (9824 per lang)