Sqft Phi 3 Mini 4k 50 Base Gptq

🚀 SQFT基礎模型：sqft-phi-3-mini-4k-50-base-gptq

SQFT基礎模型是基於特定稀疏和量化方法對源模型進行處理得到的，可用於自然語言處理等相關領域，降低模型成本並提高效率。

🚀 快速開始

若要獲取此模型，請參考 SQFT/run_command/phi-3-mini-4k-instruct/sparse_quantization.sh 中的命令。

✨ 主要特性

• 源模型：microsoft/Phi-3-mini-4k-instruct
• 稀疏方法：Wanda
• 稀疏率：50%
• 量化方法：GPTQ-INT4

📚 詳細文檔

模型來源

• 倉庫地址：https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/tree/main/SQFT
• 相關論文：
  • SQFT: Low-cost Model Adaptation in Low-precision Sparse Foundation Models
  • Low-Rank Adapters Meet Neural Architecture Search for LLM Compression

引用信息

@inproceedings{munoz-etal-2024-sqft,
    title = "{SQFT}: Low-cost Model Adaptation in Low-precision Sparse Foundation Models",
    author = "Munoz, Juan Pablo  and
      Yuan, Jinjie  and
      Jain, Nilesh",
    editor = "Al-Onaizan, Yaser  and
      Bansal, Mohit  and
      Chen, Yun-Nung",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
    month = nov,
    year = "2024",
    address = "Miami, Florida, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.findings-emnlp.749",
    pages = "12817--12832",
}

致謝

感謝稀疏算法 Wanda 和量化方法 GPTQ。

📄 許可證

本項目採用 Apache-2.0 許可證。