Update README.md

README.md

@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert-finetuned-math-prob-classification
 
-This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the competition_math dataset. Specifically, it was trained as a multi-class multi-label model.
+This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the competition_math dataset. Specifically, it was trained as a multi-class multi-label model on the problem text.
 
 ## Model description