Update README.md

README.md

@@ -7,18 +7,13 @@ model-index:
   results: []
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
+# Austin MeDeBERTa
 
-# deberta-pretrained-large
-
-This model is a fine-tuned version of [microsoft/deberta-base](https://huggingface.co/microsoft/deberta-base) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.6471
+This model was developed using further MLM pre-training on [microsoft/deberta-base](https://huggingface.co/microsoft/deberta-base), using a dataset of 1.1M clinical notes from the Austin Health EMR. The notes span discharge summaries, inpatient notes, radiology reports and histopathology reports.
 
 ## Model description
 
-More information needed
+This is the base version of the original DeBERTa model. The architecture and tokenizer are unchanged. 
 
 ## Intended uses & limitations