Update README.md

README.md

@@ -3,9 +3,9 @@ language:
 - en
 ---
 
-This dataset is the qed_train.tsv file from the FLAN project converted into alpaca formatting with the context in the input and the outputs ordered to be explanation then answer. The script is available if you would like to modify it.
-
+This dataset is the qed_train.tsv file from the FLAN project converted into alpaca formatting with the context in the input and the outputs ordered to be explanation then answer.
 
+https://github.com/google-research/FLAN/blob/main/flan/v2/cot_data/qed_train.tsv
 
 @inproceedings{weifinetuned,
   title={Finetuned Language Models are Zero-Shot Learners},