---
language:
  - en
license: apache-2.0
pipeline_tag: fill-mask
---


`Bioformer-36L` is a large bioformer model. It has 36 layers (transformer blocks) with a hidden embedding size of 1024, and the number of self-attention heads is 16.

`Bioformer-36L` has better performance than smaller models (e.g. `Bioformer-8L` and `Bioformer-16L`)