Update README.md

README.md

@@ -45,18 +45,17 @@ It achieves the following results on the evaluation set:
 - Loss: 0.1116
 - Accuracy: 0.9823
 
-## Model description
+## Base Model description
 
-More information needed
+This model is a distilled version of the [RoBERTa-base model](https://huggingface.co/roberta-base). It follows the same training procedure as [DistilBERT](https://huggingface.co/distilbert-base-uncased).
+The code for the distillation process can be found [here](https://github.com/huggingface/transformers/tree/master/examples/distillation).
+This model is case-sensitive: it makes a difference between English and English.
 
-## Intended uses & limitations
-
-More information needed
+The model has 6 layers, 768 dimension and 12 heads, totalizing 82M parameters (compared to 125M parameters for RoBERTa-base).
+On average DistilRoBERTa is twice as fast as Roberta-base.
 
 ## Training and evaluation data
 
-More information needed
-
 ## Training procedure
 
 ### Training hyperparameters