Update app.py - rename interface from Flirt-AI and Phi-3-Mini to thoshan_Flash

app.py

@@ -4,8 +4,8 @@ import gradio as gr
 import json
 import os
 
-# Load tokenizer and model - using Phi-3-mini which is publicly accessible
-model_name = "microsoft/Phi-3-mini-4k-instruct"
+# Load tokenizer and model - using thoshan_Flash model
+model_name = "microsoft/Phi-3-mini-4k-instruct"  # Will be replaced with actual thoshan_Flash model when available
 tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
 
 # Load base model directly (no LoRA adapters)
@@ -48,8 +48,8 @@ def generate_response(prompt, max_new_tokens=100):
             elif 'text' in entry:
                 context_text += f"{entry['text']}\n\n"
         context = f"Dataset context:\n{context_text}\n" if context_text else ""
-        
-    # Format the prompt for Phi-3
+    
+    # Format the prompt for thoshan_Flash
     formatted_prompt = f"<|user|>\n{context}{prompt}<|end|>\n<|assistant|>\n"
     
     inputs = tokenizer(formatted_prompt, return_tensors="pt")
@@ -63,7 +63,7 @@ def generate_response(prompt, max_new_tokens=100):
             top_p=0.9,
             pad_token_id=tokenizer.eos_token_id
         )
-        
+    
     # Decode only the generated part (excluding the input)
     generated_text = tokenizer.decode(outputs[0][inputs['input_ids'].shape[1]:], skip_special_tokens=True)
     return generated_text.strip()
@@ -76,8 +76,8 @@ iface = gr.Interface(
         gr.Slider(minimum=10, maximum=200, value=100, label="Max New Tokens")
     ],
     outputs=gr.Textbox(label="AI Response"),
-    title="Flirt-AI with Phi-3-Mini (Updated with JSONL Dataset)",
-    description="Chat with AI powered by the new flirt_dataset.jsonl dataset!"
+    title="thoshan_Flash (Updated with JSONL Dataset)",
+    description="Chat with AI powered by thoshan_Flash and the new flirt_dataset.jsonl dataset!"
 )
 
 if __name__ == "__main__":