Update README.md

README.md

@@ -18,10 +18,6 @@ This dataset contains **full** ArXiv papers randomly sampled from the train (mem
 As such, the dataset can be used to develop and evaluate document-level MIAs against LLMs trained on The Pile. 
 
 We randomly sample 1,000 documents from the train set (members) and 1,000 documents from the test set (non-members), ensuring that the selected documents have at least 5,000 words (any sequences of characters seperated by a white space). 
+We also provide the dataset where each document is split into 25 sequences of 200 words [here](https://huggingface.co/datasets/imperial-cpg/pile_arxiv_doc_mia_sequences)
 
-Target models include the suite of Pythia and GPTNeo models, to be found [here](https://huggingface.co/EleutherAI). 
-
-Note: our understanding is that the deduplication executed on the Pile to create the "Pythia-dedup" models has been only done on the training dataset, suggesting this dataset of members/non-members also to be valid for these models. 
-
-
-[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+Target models include the suite of Pythia and GPTNeo models, to be found [here](https://huggingface.co/EleutherAI). Our understanding is that the deduplication executed on the Pile to create the "Pythia-dedup" models has been only done on the training dataset, suggesting this dataset of members/non-members also to be valid for these models.