Update README.md

README.md

@@ -29,16 +29,17 @@ It achieves the following results on the evaluation set:
 
 ## Model description
 
-More information needed
+This model is a fine-tuned version of distilbert-base-uncased for text classification.
+DistilBERT encoder with a linear classification head
 
 ## Intended uses & limitations
 
-More information needed
+Direct use: quick baseline for short English text classification (binary or multiclass, per your label map).
+Out-of-scope: high-stakes decisions, long-context documents without chunking, non-English domains without re-training.
+Limitations: Augmented data is very similar original data leading to biases and lack of diversity as originally expected.
 
 ## Training and evaluation data
 
-More information needed
-
 ## Training procedure
 
 ### Training hyperparameters