Update README.md

README.md

@@ -51,4 +51,29 @@ decoded_output = tokenizer.decode(output_ids[0], skip_special_tokens=True)
 print(decoded_output)
 ```
 
-Once upon a time, the world is changing.
+Once upon a time, the world is changing.
+
+
+```
+# Now, you can use the model and tokenizer for inference
+input_text = "The Fulton County Grand Fair was set for Friday at"
+inputs = tokenizer(input_text, return_tensors="pt").to('cuda')
+
+# Generate output tokens using the model with repetition controls
+output_ids = model.generate(
+    **inputs,
+    max_length=256,  # Max tokens to generate
+    repetition_penalty=1.2,  # Penalize repeated words
+    no_repeat_ngram_size=3,  # Prevent 3-gram repetitions
+    temperature=0.9,  # Adjust randomness (lower means more deterministic)
+    top_k=50,  # Only sample from top 50 tokens
+    top_p=0.9  # Use nucleus sampling to control diversity
+)
+
+# Decode the generated token IDs back into text
+decoded_output = tokenizer.decode(output_ids[0], skip_special_tokens=True)
+
+# Print the generated output text
+print(decoded_output)
+```
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/62578ad28c6638f8a93e8856/dpDosrj8gUt2puqx5TLt_.png)