Improve dataset card metadata and formatting

README.md

@@ -1,7 +1,24 @@
+---
+task_categories:
+- text-generation
+tags:
+- question-answering
+- language-modeling
+- benchmark
+- babi
+---
 
-For the bAbI as used in [Scaling Data-Constrained Language Models](https://arxiv.org/abs/2305.16264) use commit e332ae8a626bb17178026dd14797abb9da31376e
+# Dataset Card for bAbI
 
-Creation (Copied & adapted from https://github.com/stanford-crfm/helm/blob/0eaaa62a2263ddb94e9850ee629423b010f57e4a/src/helm/benchmark/scenarios/babi_qa_scenario.py):
+This repository contains the bAbI dataset, as used in the paper:
+
+[**Scaling Data-Constrained Language Models**](https://arxiv.org/abs/2305.16264)
+
+The bAbI dataset is a set of 20 tasks designed to test the capabilities of text understanding and reasoning systems. Each task focuses on a different aspect of language understanding. For the specific version of bAbI used in the paper, please refer to commit `e332ae8a626bb17178026dd14797abb9da31376e`.
+
+## Creation / Sample Usage
+
+The data can be processed and prepared using the following Python code (copied & adapted from https://github.com/stanford-crfm/helm/blob/0eaaa62a2263ddb94e9850ee629423b010f57e4a/src/helm/benchmark/scenarios/babi_qa_scenario.py):
 
 ```python
 !wget http://www.thespermwhale.com/jaseweston/babi/tasks_1-20_v1-2.tar.gz
@@ -45,9 +62,10 @@ for split in splits:
                             "question": question,
                             "answer": answer,
                             "task": task,
-                        }) + "\n")
+                        }) + "
+")
                         if "?" in story:
                             print("STORY", "".join(story))
                     else:
                         story.append(fact)   
-```
+```