README.md

---
library_name: transformers
license: mit
pipeline_tag: text-generation
tags:
- biology
- genomics
- long-context
---

# GENERator-eukaryote-3b-base model

## **Important Notice**
If you are using GENERator for **sequence generation**, please ensure that the length of each input sequence is a multiple of **6**. This can be achieved by either:  
1. Padding the sequence on the left with `'A'` (**left padding**);  
2. Truncating the sequence from the left (**left truncation**).  

This requirement arises because **GENERator** employs a 6-mer tokenizer. If the input sequence length is not a multiple of **6**, the tokenizer will append an `'<oov>'` (out-of-vocabulary) token to the end of the token sequence. This can result in uninformative subsequent generations, such as repeated `'AAAAAA'`.  

We apologize for any inconvenience this may cause and recommend adhering to the above guidelines to ensure accurate and meaningful generation results.


## Abouts
In this repository, we present GENERator, a generative genomic foundation model featuring a context length of 98k base pairs and 3B parameters, trained on an expansive dataset comprising 386 billion base pairs of eukaryotic DNA. The extensive and diverse pre-training data endow the GENERator with enhanced understanding and generation capabilities across various organisms.

For more technical details, please refer to our paper [GENERator: A Long-Context Generative Genomic Foundation Model](https://arxiv.org/abs/2502.07272). The code and implementation details are available on Github: [https://github.com/GenerTeam/GENERator](https://github.com/GenerTeam/GENERator).


## How to use
### Simple example1:  generation

```python

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

# Load the tokenizer and model.
tokenizer = AutoTokenizer.from_pretrained("GenerTeam/GENERator-eukaryote-3b-base", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("GenerTeam/GENERator-eukaryote-3b-base")
config = model.config

max_length = config.max_position_embeddings

# Define input sequences.
sequences = [
    "ATGAGGTGGCAAGAAATGGGCTAC",
    "GAATTCCATGAGGCTATAGAATAATCTAAGAGAAAT"
]

def left_padding(sequence, padding_char='A', multiple=6):
    remainder = len(sequence) % multiple
    if remainder != 0:
        padding_length = multiple - remainder
        return padding_char * padding_length + sequence
    return sequence

def left_truncation(sequence, multiple=6):
    remainder = len(sequence) % multiple
    if remainder != 0:
        return sequence[remainder:]
    return sequence

# Apply left_padding to all sequences
# padded_sequences = [left_padding(seq) for seq in sequences]

# Apply left_truncation to all sequences
truncated_sequences = [left_truncation(seq) for seq in sequences]

# Process the sequences
sequences = [tokenizer.bos_token + sequence for sequence in truncated_sequences]

# Tokenize the sequences
tokenizer.padding_side = "left"
inputs = tokenizer(
    sequences,
    add_special_tokens=False,
    return_tensors="pt",
    padding=True,
    truncation=True,
    max_length=max_length
)

# Generate the sequences
with torch.inference_mode():
    outputs = model.generate(**inputs, max_new_tokens=32, temperature=0.00001, top_k=1)

# Decode the generated sequences
decoded_sequences = tokenizer.batch_decode(outputs, skip_special_tokens=True)

# Print the decoded sequences
print(decoded_sequences)

# It is expected to observe non-sense decoded sequences (e.g., 'AAAAAA')
# The input sequences are too short to provide sufficient context.
```

### Simple example2: embedding

```python

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

# Load the tokenizer and model.
tokenizer = AutoTokenizer.from_pretrained("GENERator-eukaryote-3b-base", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("GenerTeam/GENERator-eukaryote-3b-base")

config = model.config
max_length = config.max_position_embeddings

# Define input sequences.
sequences = [
    "ATGAGGTGGCAAGAAATGGGCTAC",
    "GAATTCCATGAGGCTATAGAATAATCTAAGAGAAAT"
]

# Tokenize the sequences with add_special_tokens=True to automatically add special tokens,
# such as the BOS EOS token, at the appropriate positions.
tokenizer.padding_side = "right"
inputs = tokenizer(
    sequences,
    add_special_tokens=True,
    return_tensors="pt",
    padding=True,
    truncation=True,
    max_length=max_length
)

# Perform a forward pass through the model to obtain the outputs, including hidden states.
with torch.inference_mode():
    outputs = model(**inputs, output_hidden_states=True)

# Retrieve the hidden states from the last layer.
hidden_states = outputs.hidden_states[-1]  # Shape: (batch_size, sequence_length, hidden_size)

# Use the attention_mask to determine the index of the last token in each sequence.
# Since add_special_tokens=True is used, the last token is typically the EOS token.
attention_mask = inputs["attention_mask"]
last_token_indices = attention_mask.sum(dim=1) - 1  # Index of the last token for each sequence

# Extract the embedding corresponding to the EOS token for each sequence.
seq_embeddings = []
for i, token_index in enumerate(last_token_indices):
    # Fetch the embedding for the last token (EOS token).
    seq_embedding = hidden_states[i, token_index, :]
    seq_embeddings.append(seq_embedding)

# Stack the embeddings into a tensor with shape (batch_size, hidden_size)
seq_embeddings = torch.stack(seq_embeddings)

print("Sequence Embeddings:", seq_embeddings)

```

## Citation
```
@misc{wu2025generator,
      title={GENERator: A Long-Context Generative Genomic Foundation Model}, 
      author={Wei Wu and Qiuyi Li and Mingyang Li and Kun Fu and Fuli Feng and Jieping Ye and Hui Xiong and Zheng Wang},
      year={2025},
      eprint={2502.07272},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2502.07272}, 
}
```