Update app.py

app.py

@@ -7,7 +7,7 @@ import gradio as gr
 import sentencepiece
 
 title = "Welcome to Tonic's 🐋🐳Orca-2-13B!"
-description = "You can use [🐋🐳microsoft/Orca-2-13b](https://huggingface.co/microsoft/Orca-2-13b) via API using Gradio by scrolling down and clicking Use 'Via API' or privately by [cloning this space on huggingface](https://huggingface.co/spaces/Tonic1/TonicsOrca2?duplicate=true) . [Join me on Discord to build together](https://discord.gg/VqTxc76K3u). Big thanks to the HuggingFace Organisation for the Community Grant."
+description = "You can use [🐋🐳microsoft/Orca-2-13b](https://huggingface.co/microsoft/Orca-2-13b) via API using Gradio by scrolling down and clicking Use 'Via API' or privately by [cloning this space on huggingface](https://huggingface.co/spaces/Tonic1/TonicsOrca2?duplicate=true) . [Join my active builders' server on discord](https://discord.gg/VqTxc76K3u). Big thanks to the HuggingFace Organisation for the Community Grant."
 
 # os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'max_split_size_mb:50'
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
@@ -45,8 +45,10 @@ class OrcaChatBot:
             temperature=temperature,
             top_p=top_p,
             repetition_penalty=repetition_penalty,
-            pad_token_id=self.tokenizer.eos_token_id
+            pad_token_id=self.tokenizer.eos_token_id,
+            do_sample=True  # Enable sampling-based generation
         )
+        
 
         # Decode the generated response
         response = self.tokenizer.decode(output_ids[0], skip_special_tokens=True)