Update README.md

README.md

@@ -52,7 +52,7 @@ print(label_names[prediction.argmax(0).tolist()])
 
 ### Training data
 
-This model was trained on the MultiNLI dataset, which consists of 433k sentence textual entitlement. 
+This model was trained on the MultiNLI dataset, which consists of 392K sentence textual entitlement. 
 
 ### Training procedure
 
@@ -82,10 +82,6 @@ train_args = TrainingArguments(
 
 ```
 
-## Limitations and bias
-
-Please consult the original DeBERTa paper and literature on different NLI datasets for potential biases. 
-
 ### BibTeX entry and citation info
 
 Please cite the [DeBERTa paper](https://arxiv.org/abs/2006.03654) and [MultiNLI Dataset](https://cims.nyu.edu/~sbowman/multinli/paper.pdf) if you use this model and include the Huggingface hub.