Update README.md

README.md

@@ -1,24 +1,42 @@
 ---
 license: mit
+tags:
+- gpt2
+- distilled
+- small-llm
+- llama3
+- alpaca
+- reasoning
+- chatbot
+- conversational
 ---
-**Lazarus, the best small LLM**
 
-Lazarus is a next-generation LLM based off of GPT2-medium and distilled from Llama3 3b using the vicgalle/alpaca-gpt4 dataset.
+# Lazarus — The Best Small LLM
 
-Lazarus shows exceptional understanding of questions and reasoning compared to other small models its size. Much better, larger model sizes are coming.
+**Lazarus** is a next-generation small LLM based on `gpt2-medium`, distilled from LLaMA 3 3B using the [`vicgalle/alpaca-gpt4`](https://huggingface.co/datasets/vicgalle/alpaca-gpt4) dataset.
 
-Here are some of Lazarus's responses:
+Lazarus demonstrates exceptional performance for its size, especially in question understanding and step-by-step reasoning — outperforming many small LLMs. Larger versions are in development.
 
-User: what is AI?
-Lazarus: Artificial Intelligence (AI) is a technology that helps humans solve complex problems and communicate with each other. It can help humans solve complex problems by analyzing and modeling data, and can help them understand complex information.
+---
+
+## 🔍 Example Responses
+
+**User:** What is AI?  
+**Lazarus:** Artificial Intelligence (AI) is a technology that helps humans solve complex problems and communicate with each other. It can help humans solve complex problems by analyzing and modeling data, and can help them understand complex information.
+
+**User:** What is the Pythagorean Theorem?  
+**Lazarus:** The Pythagorean theorem is a theorem used to find the hypotenuse of a triangle.
 
-User: what is the Pythagorean Theorem?
-Lazarus: The pythagorean theorem is a theorem used to find the hypotenuse of a triangle.
+---
+
+## 🚀 Usage
 
-Here is the code to run:
+You can interact with Lazarus using the script below:
 
-python3 ''' 
+<details>
+<summary>Click to expand Python code</summary>
 
+```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
 import torch
 
@@ -47,10 +65,7 @@ def chat():
         if user_input.lower() == "exit":
             break
 
-        # Add user's message
         chat_history.append(f"You: {user_input}")
-
-        # Keep last 3 exchanges (6 lines)
         recent_history = chat_history[-6:]
         full_prompt = system_prompt + "\n".join(recent_history) + "\nAI:"
 
@@ -68,11 +83,9 @@ def chat():
                 eos_token_id=tokenizer.eos_token_id
             )
 
-        # Extract model's new output portion
         response = tokenizer.decode(outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True)
         response = response.strip()
 
-        # Filter out low-quality or empty responses
         bad_responses = {"I hope that", "I don't know", "", "I'm excited"}
         if response in bad_responses:
             print("AI: [Regenerating due to low-quality response]")
@@ -83,5 +96,3 @@ def chat():
 
 if __name__ == "__main__":
     chat()
-
-'''