Doge 20M Instruct

🚀 Doge 20M Instruct

Doge 20M Instruct uses Dynamic Mask Attention for sequence transformation and can leverage Multi - Layer Perceptron or Cross Domain Mixture of Experts for state transformation. It is trained by the SmallDoge community.

🚀 Quick Start

Doge uses Dynamic Mask Attention as sequence transformation and can use Multi - Layer Perceptron or Cross Domain Mixture of Experts as state transformation. Dynamic Mask Attention allows the Transformer to use self - attention during training and state space during inference, and Cross Domain Mixture of Experts can directly inherit the weights of Multi - Layer Perceptron for further training. This model is trained by SmallDoge community. For detailed algorithm and model architecture, a paper is coming soon. All training details and code are available in the small - doge repository.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig, TextStreamer

tokenizer = AutoTokenizer.from_pretrained("SmallDoge/Doge-20M-Instruct")
model = AutoModelForCausalLM.from_pretrained("SmallDoge/Doge-20M-Instruct", trust_remote_code=True)

generation_config = GenerationConfig(
      max_new_tokens=100, 
      use_cache=True, 
      do_sample=True, 
      temperature=0.8, 
      top_p=0.9,
      repetition_penalty=1.0
)
steamer = TextStreamer(
      tokenizer=tokenizer, 
      skip_prompt=True
)

prompt = "Hi, how are you doing today?"
conversation = [
      {"role": "user", "content": prompt}
]
inputs = tokenizer.apply_chat_template(
    conversation=conversation,
    tokenize=True,
    return_tensors="pt",
)

outputs = model.generate(
    inputs, 
    tokenizer=tokenizer,
    generation_config=generation_config, 
    streamer=steamer
)

📚 Documentation

Model Details

We build the Doge - Instruct by first SFT on SmolTalk and then DPO on UltraFeedback Binarized.

SFT:

DPO:

Evaluation:

Procedure: SFT:

DPO:

Environment:

• Image: nvcr.io/nvidia/pytorch:24.12 - py3
• Hardware: 1x NVIDIA RTX 4090
• Software: Transformers, TRL

📄 License

This project is licensed under the Apache - 2.0 license.

📖 Citation

@misc{smalldoges,
  title={SmallDoges: A Family of Dynamic UltraFast Small Language Models}, 
  author={Jingze, Shi and Yifan, Wu and Bingheng, Wu and Yuyu, Luo},
  year={2025},
  month={March},
  url={https://github.com/SmallDoges/small-doge}
}