.

README.md

@@ -44,27 +44,31 @@ The model utilizes a custom implementation of the Gemma3 architecture:
 - **Hardware:** Single NVIDIA A100 GPU (40GB).
 - **Development Context:** This project was developed at **Tunica Tech** as a case study in Small Language Model (SLM) alignment and Reinforcement Learning.
 
-## Usage
+# Requirements
+pip install git+https://huggingface.co/Shubhamw11/Gemma-270M-TinyStories
 
-To use this model, ensure `modeling_gemma3.py` is present in your working directory.
+
+## How to use
 
 ```python
-from transformers import AutoModelForCausalLM
-import torch
+from gemma3_tinystories import HFGemma3DPONegative, Gemma3Config
 import tiktoken
+import torch
 
 # Load Aligned Model
-model_id = "shubham-waghmare/gemma-3-tinystories-negative-dpo"
-model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True)
+config = Gemma3Config.from_pretrained("Shubhamw11/gemma-3-270m-dpo-negative")
+model = HFGemma3DPONegative.from_pretrained("Shubhamw11/gemma-3-270m-dpo-negative", config=config).model
 tokenizer = tiktoken.get_encoding("gpt2")
+```
 
+## Generate text
+```python
 device = "cuda" if torch.cuda.is_available() else "cpu"
-model.to(device)
 
-# Generation
-prompt = "Once upon a time, there was a guy named Max,"
-input_ids = torch.tensor(tokenizer.encode(prompt)).unsqueeze(0).to(device)
+input_text = "Once upon a time, there was a little"
+context = torch.tensor(tokenizer.encode(input_text), dtype=torch.long).unsqueeze(0).to(device)
+model.to(device)
+response = model.generate(context, max_new_tokens=200, temperature=1.1, top_k=5)
 
-output = model.generate(input_ids, max_new_tokens=150, temperature=1.0)
-print(tokenizer.decode(output.squeeze().tolist()))
+print(tokenizer.decode(response.squeeze().tolist()))
 ```