Add model card

README.md

@@ -0,0 +1,13 @@
+---
+license: mit
+library_name: transformers
+pipeline_tag: feature-extraction
+---
+
+# EvoLen DNA Language Model
+
+This is a BERT-based masked language model pretrained on the human genome using the EvoLen tokenizer described in the paper [EvoLen: Evolution-Guided Tokenization for DNA Language Model](https://huggingface.co/papers/2604.08698). The tokenizer incorporates evolutionary conservation signals to better preserve functional sequence patterns.
+
+The model can be used for feature extraction from DNA sequences and fine-tuning on downstream genomics tasks.
+
+Code for tokenizer construction, pretraining, and evaluation is available at [https://github.com/HN020719/EvoLen](https://github.com/HN020719/EvoLen).