Update app.py

app.py

@@ -50,30 +50,24 @@ Description: """
         if tokenizer.pad_token_id is None:
             tokenizer.pad_token_id = tokenizer.eos_token_id
         
-        # Tokenize input with explicit token IDs
+        # Tokenize input
         inputs = tokenizer(
             input_text,
             return_tensors="pt",
-            padding=True,
-            truncation=True,
             add_special_tokens=True
         )
         
-        input_length = inputs['input_ids'].shape[1]
-        
-        # Generate response
+        # Generate response with simpler parameters
         with torch.no_grad():
             outputs = model.generate(
                 input_ids=inputs['input_ids'],
-                attention_mask=inputs['attention_mask'],
-                max_new_tokens=max_length,  # Changed to max_new_tokens
+                max_new_tokens=max_length,
+                do_sample=True,
                 temperature=temperature,
                 top_p=top_p,
-                do_sample=True,
-                num_return_sequences=1,
+                use_cache=False,  # Disable caching to avoid the error
                 pad_token_id=tokenizer.pad_token_id,
-                eos_token_id=tokenizer.eos_token_id,
-                use_cache=True
+                eos_token_id=tokenizer.eos_token_id
             )
         
         # Decode and return the response