Update README.md

README.md

@@ -8,7 +8,7 @@ base_model:
 - MedSwin/MedSwin-7B-KD
 pipeline_tag: question-answering
 metrics:
-- bertscore: 0.8341
+- bertscore: 0.8297
 tags:
 - medical
 ---
@@ -19,6 +19,6 @@ tags:
 - **License:** Apache 2.0
   
 ## Model Details
-This is a Supervised Finetuning (SFT) model on top of a Knowledge Distillation model (KD), on synethised dataset from PubMedQA.
+This is a Supervised Finetuning (SFT) model on top of a Knowledge Distillation model (KD), on synethised-mix dataset from PubMedQA.
 
 > Review all model metrics benchmark via [Benchmark Document Preview](https://hackmd.io/@ngFNmXW1RVOfNb7b3NYBJg/model_review).