Update app.py

app.py

@@ -2,6 +2,7 @@ import gradio as gr
 from transformers import AutoTokenizer, AutoModelForCausalLM
 import torch
 
+# Load the model
 model_path = "BytArch/source-mini"
 tokenizer = AutoTokenizer.from_pretrained(model_path)
 model = AutoModelForCausalLM.from_pretrained(model_path)
@@ -9,6 +10,7 @@ model = AutoModelForCausalLM.from_pretrained(model_path)
 if tokenizer.pad_token is None:
     tokenizer.pad_token = tokenizer.eos_token
 
+# System prompt
 SYSTEM_PROMPT = (
     "You are source-mini-beta, an open-source chatbot developed by BytArch and created by Joshua Kelly. "
     "You are currently in beta development and are still learning and improving. "
@@ -28,9 +30,11 @@ SYSTEM_PROMPT = (
     "Now, greet the user and let them know you are ready to chat in whichever style they prefer."
 )
 
+# Build context for stateless generation
 def build_context(user_message):
     return SYSTEM_PROMPT + "\n\nUser: " + user_message + "\nAssistant:"
 
+# Generate response
 def generate_response(
     prompt,
     max_tokens=300,
@@ -68,8 +72,10 @@ def generate_response(
 
     return response.strip()
 
+# Respond function for Gradio
 def respond(
     message,
+    history,  # required by ChatInterface but ignored
     max_tokens,
     temperature,
     top_p,
@@ -85,6 +91,7 @@ def respond(
         top_k=top_k,
     )
 
+# Gradio interface
 chatbot = gr.ChatInterface(
     respond,
     type="messages",