Update Usage

README.md

@@ -33,9 +33,10 @@ The dataset includes Japanese instruction-response pairs and has been tailored f
 
 ## Usage
 
-1. Install Required Libraries
 
-```
+### 1. Install Required Libraries
+
+```python
 !pip install -U bitsandbytes
 !pip install -U transformers
 !pip install -U accelerate
@@ -43,9 +44,9 @@ The dataset includes Japanese instruction-response pairs and has been tailored f
 !pip install -U peft
 ```
 
-2. Load the Model and Libraries
+### 2. Load the Model and Libraries
 
-```
+```python
 from transformers import (
     AutoModelForCausalLM,
     AutoTokenizer,
@@ -69,9 +70,9 @@ bnb_config = BitsAndBytesConfig(
 )
 ```
 
-3. Load the Base Model and LoRA Adapter
+### 3. Load the Base Model and LoRA Adapter
 
-```
+```python
 # Load base model with 4-bit quantization
 model = AutoModelForCausalLM.from_pretrained(
     base_model_id,
@@ -91,9 +92,9 @@ tokenizer = AutoTokenizer.from_pretrained(
 model = PeftModel.from_pretrained(model, adapter_id, token=HF_TOKEN)
 ```
 
-4. Perform Inference
+### 4. Perform Inference
 
-```
+```python
 # Example input prompt
 input_text = """次の文章を要約してください。