Create README.md

README.md

@@ -0,0 +1,14 @@
+## Generation procedure
+
+The dataset was constructed using documents from [the Pile](https://pile.eleuther.ai/) scored using using [Perspective API](http://perspectiveapi.com) toxicity scores.
+
+The procedure was the following:
+1. A chunk of the Pile (3%, 7m documents) was scored using the Perspective API.
+1. The first half of this dataset is [tomekkorbak/pile-toxic-chunk-0](https://huggingface.co/datasets/tomekkorbak/pile-toxic-chunk-0), 100k *most* toxic documents of the scored chunk
+2. The first half of this dataset is [tomekkorbak/pile-nontoxic-chunk-0](https://huggingface.co/datasets/tomekkorbak/pile-nontoxic-chunk-0), 100k most *least* documents of the scored chunk
+3. Then, the dataset was shuffled and a 9:1 train-test split was done
+
+## Basic stats
+
+The average scores of the good and bad half are 0.0014 and 0.67, respectively. The average score of the whole dataset is 0.33; the median is 0.51.
+