Update README.md

README.md

@@ -76,8 +76,9 @@ Java Subset:
 - Security: Code suggestions should not be assumed to be secure or free of vulnerabilities.
 
 # Sample Usage
-Here’s an example of how to run and sample from the model:
+Here are examples of how to run and sample from the model.
 
+## Generic generaion
 ```python
 import json
 from transformers import AutoTokenizer, AutoModelForCausalLM
@@ -129,6 +130,25 @@ print("### Prediction")
 print(tokenizer.decode(out[0][input_len:]))
 ```
 
+## Fill in the middle generation
+```python
+prefix = """
+def fibonacci(n: int) -> int:
+"""
+
+suffix = """
+if __name__ == "__main__":
+    print(fibonacci(10))
+"""
+
+encoded_input = tokenizer(f"<fim_suffix>suffix<fim_prefix>{prefix}<fim_middle>", return_tensors='pt', return_token_type_ids=False)
+input_len = len(encoded_input["input_ids"][0])
+out = model.generate(
+    **encoded_input,
+    max_new_tokens=100,
+)
+```
+
 # Citation
 If you use this model, please cite: