datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
gptneo-pubmed-abstracts
Dataset Card for Dataset Name
This dataset card aims to be a base template for new datasets. It has been generated using this raw template.
Dataset Details
Dataset Description
This is a dataset consisting of 10000 PubMed abstracts from The Pile (arXiv:2101.00027), along with completions (both human, and LLM-generated), in order to be used to calculate Heaps Law, in the manner described in the preliminary paper, Heaps' Law in GPT-Neo Large Language Model… See the full description on the dataset page: https://huggingface.co/datasets/rachel6603/gptneo-pubmed-abstracts.PubMed-Mimic-Gpt-Neoheheheh