---
library_name: transformers
tags:
- bert
- language-model
license: apache-2.0
---

# BERT-updated

Standard BERT architecture with `flash_attention_2` and `sdpa` support added.

This is a **shared code repository** — it contains no pretrained weights. It is used
as the code backend for biological sequence models that share the vanilla BERT
architecture (post-LN transformer, learned absolute position embeddings) but have
model-specific vocabularies and hyperparameters:

- [Taykhoom/RNABERT](https://huggingface.co/Taykhoom/RNABERT)
- [Taykhoom/UTRBERT-3mer](https://huggingface.co/Taykhoom/UTRBERT-3mer), [4mer](https://huggingface.co/Taykhoom/UTRBERT-4mer), [5mer](https://huggingface.co/Taykhoom/UTRBERT-5mer), [6mer](https://huggingface.co/Taykhoom/UTRBERT-6mer)
- [Taykhoom/DNABERT-3mer](https://huggingface.co/Taykhoom/DNABERT-3mer), [4mer](https://huggingface.co/Taykhoom/DNABERT-4mer), [5mer](https://huggingface.co/Taykhoom/DNABERT-5mer), [6mer](https://huggingface.co/Taykhoom/DNABERT-6mer)

Each of those repos stores weights, tokenizer, and config; their `auto_map` in
`config.json` points here for the modeling code.

## What was changed from stock `transformers.BertModel`

The standard HF `BertModel` (transformers 4.57.6) supports `sdpa` but not
`flash_attention_2`. This repo adds a complete `attn_implementation` dispatch:

| Backend | Class | Notes |
|---|---|---|
| `eager` | `BertSelfAttention` | Standard scaled dot-product, identical to original BERT |
| `sdpa` | `BertSdpaSelfAttention` | `F.scaled_dot_product_attention`, bool mask -> additive float mask |
| `flash_attention_2` | `BertFlashSelfAttention` | `flash_attn_varlen_func` for padded inputs, `flash_attn_func` for unpadded |

The rest of the architecture (embeddings, FFN, pooler, weight layout) is unchanged.

## Usage

Do not load this repo directly. Load one of the model repos listed above:

```python
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("Taykhoom/RNABERT", trust_remote_code=True)
model = AutoModel.from_pretrained("Taykhoom/RNABERT", trust_remote_code=True)

# Flash Attention 2
model = AutoModel.from_pretrained("Taykhoom/UTRBERT-3mer", trust_remote_code=True,
                                   attn_implementation="flash_attention_2")
```

## Credits

Modeling code authored primarily by [Claude Code](https://claude.ai/code) and reviewed
manually by Taykhoom Dalal.

## License

Apache 2.0.