improved app

Pretrained_GPT_med_bot.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce9906732c175852c397380acc7298cd42fe5019b3744faef55d04d3536096eb
-size 128625029
+oid sha256:a58d2701e932974d8090e3fba87fef031ec7b85a090f0a61d218e2783a876ec4
+size 128627717

app.py

@@ -28,7 +28,7 @@ def main():
         model.eval()
         with torch.inference_mode():
             response = model.generate(prompt.unsqueeze(0),
-                                      max_new_tokens=200)[0].tolist()
+                                      max_new_tokens=100)[0].tolist()
         answer = decode(response)[in_len:]
 
         pred_time = round(timer() - start, 5)