Update README.md

README.md

@@ -51,21 +51,24 @@ task_ids:
 - **Size of downloaded dataset files:** 
 - **Size of the generated dataset:** 
 - **Total amount of disk used:**
-### Dataset usage
-Available "kneset16","kneset17","knesset_tagged" configurations
-And only train set.
-```python
-train_ds = load_dataset("imvladikon/knesset_meetings_corpus", "kneset16", split="train")
-```
-### Data Instances
-An example of sample:
+### Dataset Summary
+
+An example of a sample:
 ```
 {
     "text": <text content of given document>,
     "path": <file path to docx>
 }
 ```
-### Dataset Summary
+
+Dataset usage
+Available "kneset16","kneset17","knesset_tagged" configurations
+And only train set.
+```python
+train_ds = load_dataset("imvladikon/knesset_meetings_corpus", "kneset16", split="train")
+```
+
+
 The Knesset Meetings Corpus 2004-2005 is made up of two components:
 
 * Raw texts - 282 files made up of 867,725 lines together. These can be downloaded in two formats: