Update app.py

app.py

@@ -50,14 +50,17 @@ class HfModelWrapper:
         streamer = TextIteratorStreamer(self.tokenizer, skip_prompt=True)#skip_special_tokens=True)
 
         gen_kwargs = {
-            "inputs": input_ids,
+            "input_ids": input_ids,
             "streamer": streamer,
-            "pad_token_id": self.tokenizer.eos_token_id,
-            "max_length": 32768,
-            "temperature": 0.1,
-            "top_p": 0.9,
-            #"repetition_penalty": 1.1,
+            "max_new_tokens": 2048,  
             "min_length": 10,
+            "temperature": 0.7,
+            "top_p": 0.9,
+            "top_k": 50,
+            "repetition_penalty": 1.1,
+            "do_sample": True,
+            "eos_token_id": tokenizer.eos_token_id,
+            "pad_token_id": tokenizer.pad_token_id, 
         }
 
         thread = Thread(target=self.model.generate, kwargs=gen_kwargs)