Update app.py

app.py

@@ -6,17 +6,17 @@ import torch
 import gradio as gr
 import sentencepiece
 
-title = "# ²Welcome to 🙋🏻‍♂️Tonic's🌷Tulu Chat!"
-description = """[allenai/tulu-2-dpo-7b](https://huggingface.co/allenai/tulu-2-dpo-7b) and larger Tulu-2 models are Instruct Llama Finetunes using the [mistralai/Mistral-7B](https://huggingface.co/mistralai/Mistral-7B-v0.1) recipe. You can use [allenai/tulu-2-13b](https://huggingface.co/allenai/tulu-2-13b) here 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/TuluDemo?duplicate=true) See also the large model here : [allenai/tulu-2-dpo-70b](https://huggingface.co/allenai/tulu-2-dpo-70b) . [Join my active builders' server on discord](https://discord.gg/VqTxc76K3u). Let's build together!. [Add this Space as a discord bot to your server by clicking this link](https://discord.com/oauth2/authorize?client_id=1176628808212828231&scope=bot+applications.commands&permissions=326417525824). Big thanks to 🤗Huggingface Organisation for the🫂Community Grant"""
+title = "# ²Welcome to 🙋🏻‍♂️Tonic's🧠🤌🏻Neural Chat (From Intel)!"
+description = """Try out [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1) the Instruct Llama Finetune using the [mistralai/Mistral-7B](https://huggingface.co/mistralai/Mistral-7B-v0.1) recipe. You can use [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1) here via API using Gradio by scrolling down and clicking Use 'Via API' or privately by [cloning this space on huggingface](https://huggingface.co/spaces/TeamTonic/NeuralChat?duplicate=true)  . [Join my active builders' server on discord](https://discord.gg/VqTxc76K3u). Let's build together!. """
 
 os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'max_split_size_mb:50'
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
-model_name = "allenai/tulu-2-dpo-13b"
-tokenizer = AutoTokenizer.from_pretrained("allenai/tulu-2-dpo-13b")
+model_name = "Intel/neural-chat-7b-v3-1"
+tokenizer = AutoTokenizer.from_pretrained("Intel/neural-chat-7b-v3-1")
 model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, device_map="auto")
 
-class TuluChatBot:
-    def __init__(self, model, tokenizer, system_message="You are 🌷Tulu, an AI language model created by Tonic-AI. You are a cautious assistant. You carefully follow instructions. You are helpful and harmless and you follow ethical guidelines and promote positive behavior."):
+class IntelChatBot:
+    def __init__(self, model, tokenizer, system_message="You are 🧠🤌🏻Neuro, an AI language model created by Tonic-AI. You are a cautious assistant. You carefully follow instructions. You are helpful and harmless and you follow ethical guidelines and promote positive behavior."):
         self.model = model
         self.tokenizer = tokenizer
         self.system_message = system_message
@@ -25,7 +25,7 @@ class TuluChatBot:
         self.system_message = new_system_message
 
     def format_prompt(self, user_message):
-        prompt = f"<|assistant|>\n {self.system_message}\n\n <|user|>{user_message}\n\n<|assistant|>\n"
+        prompt = f"### System:\n {self.system_message}\n ### User:\n{user_message}\n### Assistant:\n"
         return prompt
 
     def predict(self, user_message, temperature, max_new_tokens, top_p, repetition_penalty, do_sample):
@@ -49,23 +49,23 @@ class TuluChatBot:
         return response
 
 def gradio_predict(user_message, system_message, max_new_tokens, temperature, top_p, repetition_penalty, do_sample):
-    Tulu_bot.set_system_message(system_message)
+    Intel_bot.set_system_message(system_message)
     if not do_sample:
         max_length = 780
         temperature = 1.2  
         top_p = 0.9        
         repetition_penalty = 0.9  
-    response = Tulu_bot.predict(user_message, temperature, max_new_tokens, top_p, repetition_penalty, do_sample)
+    response = Intel_bot.predict(user_message, temperature, max_new_tokens, top_p, repetition_penalty, do_sample)
     return response
 
-Tulu_bot = TuluChatBot(model, tokenizer)
+Intel_bot = IntelChatBot(model, tokenizer)
 
 
 with gr.Blocks(theme = "ParityError/Anime") as demo:
     gr.Markdown(title)  
     gr.Markdown(description)  
     with gr.Row():
-        system_message = gr.Textbox(label="Optional 🌷Tulu Assistant Message", lines=2)
+        system_message = gr.Textbox(label="Optional 🧠🤌🏻NeuralChat Assistant Message", lines=2)
         user_message = gr.Textbox(label="Your Message", lines=3)
     with gr.Row():
         do_sample = gr.Checkbox(label="Advanced", value=False)
@@ -78,7 +78,7 @@ with gr.Blocks(theme = "ParityError/Anime") as demo:
             repetition_penalty = gr.Slider(label="Repetition penalty", value=1.9, minimum=1.0, maximum=2.0, step=0.05)
 
     submit_button = gr.Button("Submit")
-    output_text = gr.Textbox(label="🌷Tulu Response")
+    output_text = gr.Textbox(label="🧠🤌🏻NeuralChat Response")
 
     def process(user_message, system_message, max_new_tokens, temperature, top_p, repetition_penalty, do_sample):
         return gradio_predict(user_message, system_message, max_new_tokens, temperature, top_p, repetition_penalty, do_sample)