Update README.md

README.md

@@ -76,7 +76,7 @@ python3 evaluate.py --predicted-path predictions.txt --gold-path targets.txt --l
 
 The dataset is in Russian.
 
-## Usage
+### Usage
 
 Loading version 1.0:
 ```python
@@ -90,7 +90,7 @@ from datasets import load_dataset
 dataset = load_dataset('IlyaGusev/gazeta', script_version="v2.0")
 ```
 
-## Other datasets
+### Other datasets
 Other Russian summarization datasets:
   * Russian part of [XL-Sum](https://huggingface.co/datasets/csebuetnlp/xlsum), parsed from www.bbc.com/russian, 77803 samples
   * Russian part of [MLSUM](https://huggingface.co/datasets/mlsum), parsed from www.mk.ru, 27063 samples