EADST

Quick Review: SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression

作者：XD / 发表： 2023年12月6日 23:57 / 更新： 2023年12月6日 23:57 / 科研学习 / 阅读量：2326

SpQR: A Sparse-Quantized Representation for Near-Lossless Large Language Model Weight Compression

Core Approach:

  • GPTQ without Outliers: Focuses on eliminating outliers during the GPTQ process, enabling more efficient and accurate weight compression for large language models.

本文作者：XD 转载请标明出处：http://www.eadst.com/blog/225

本站采用知识共享署名-非商业性使用-相同方式共享 4.0 国际许可协议进行许可。

上一篇
Quick Review: Optimize Weight Rounding via Signed Gradient Descent for the Quantization of LLMs
下一篇
Quick Review: QUIK: Towards End-to-end 4-Bit Inference on Generative Large Language Models
相关标签
LLM Quantization
About Me
XD
Goals determine what you are going to be.
Category
标签云
COCO LoRA News LeetCode diffusers torchinfo InvalidArgumentError 顶会 OpenCV Qwen2.5 Tiktoken BF16 Tracking Vim Color FP16 DeepStream OpenAI Transformers Nginx Bitcoin git Quantization PyCharm 关于博主 SPIE Pillow Heatmap Crawler Gemma Input PDF 腾讯云 Interview CUDA Clash 图标 CSV Datetime SQL icon CC RGB WAN API网关 Website Excel 云服务器 Llama Card Logo Mixtral FastAPI SVR CAM Search NLTK Paddle Jupyter Password TensorRT BeautifulSoup Web Claude 多进程 Qwen 论文速读 Food Hotel FP8 Miniforge Paper uWSGI Python 版权 tar Jev Ubuntu Docker GIT PIP mmap Translation Shortcut v0.dev OCR Breakpoint Land SAM TTS 签证 FP32 API Diagram Augmentation printf LLM Algorithm Quantize GGML Distillation PDB Magnet Baidu Use XML Hungarian JSON Template SQLite Anaconda XGBoost Sklearn logger hf Pytorch UI Animate Proxy Domain Qwen2 Vmess git-lfs Video Pickle Attention Windows Bin HuggingFace Image2Text Rebuttal 第一性原理 Linux Plate Streamlit scipy Math Tensor CTC IndexTTS2 ms-swift Django Plotly 论文 MD5 FP64 飞书 RL CV Pandas Michelin PyTorch Zip 报税 搞笑 Knowledge Harness EXCEL 公式 Disk 财报 NLP 强化学习 Bert ModelScope Hilton 证件照 GoogLeNet Statistics Jetson NameSilo VSCode ONNX llama.cpp YOLO ResNet-50 TSV WebCrawler Bipartite Base64 Review ChatGPT Google LaTeX Permission 递归学习法 Firewall LLAMA FlashAttention 图形思考法 Agent 多线程 CLAP RAR VPN C++ 算法题 Git Safetensors uwsgi QWEN 音频 Cloudreve Freesound 净利润 BTC Conda Dataset VGG-16 HaggingFace transformers Data TensorFlow AI Github 域名 GPT4 Markdown Ptyhon tqdm GPTQ DeepSeek 继承 CEIR v2ray UNIX 阿里云 Numpy Random
站点统计

本站现有博文338篇,共被浏览988006次

本站已经建立2700天!

热门文章
文章归档
回到顶部