Add dataset card

README.md

@@ -0,0 +1,16 @@
+---
+license: mit
+task_categories:
+- text-classification
+tags:
+- biology
+- dna
+---
+
+# EvoLen Dataset
+
+This dataset is part of the [EvoLen: Evolution-Guided Tokenization for DNA Language Model](https://huggingface.co/papers/2604.08698) project. It contains tokenized DNA sequences and benchmark data for evaluating DNA language models, including pretraining corpora and downstream task benchmarks such as GUE, Nucleotide Transformer, GenomicBenchmarks, ATAC, and SCREEN.
+
+Code: [https://github.com/HN020719/EvoLen](https://github.com/HN020719/EvoLen)
+
+For detailed documentation regarding tokenizer construction, pretraining, and benchmark preparation, please refer to the [GitHub repository](https://github.com/HN020719/EvoLen).