LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale

作者： Tim Dettmers, Mike Lewis, Younes Belkada, Luke Zettlemoyer (2022)

arXiv： 2208.07339

领域

推理

TLDR（中文）

揭示大模型激活中的"emergent outliers"，并提出混合精度方案。bitsandbytes 库背后的核心工作，让 175B 模型第一次能塞进 8 卡 A100。

TLDR (English)

Reveals "emergent outliers" in large model activations and proposes mixed-precision solution. Core work behind bitsandbytes library, first enabling 175B models to fit in 8 A100s.

出现在这些文章里

• KV Cache 与量化：让大模型跑得更快
• KV Cache and Quantization: Making Large Models Faster

同被引用

这些论文与本文出现在同一篇文章中

• QLoRA: Efficient Finetuning of Quantized LLMs Tim Dettmers et al. (2023)
• GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers Elias Frantar et al. (2022)
• Efficient Memory Management for Large Language Model Serving with PagedAttention Woosuk Kwon et al. (2023)
• AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration Ji Lin et al. (2023)

相关论文

同一领域的其他论文

• DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model DeepSeek-AI (2024)
  架构推理
• FlashAttention-3: Fast and Accurate Attention with Asympotic IO Complexity Jay Shah et al. (2024)
  推理
• Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters Charlie Snell et al. (2024)
  推理能力推理
• GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints Joshua Ainslie et al. (2023)
  推理架构
• Accelerating Large Language Model Decoding with Speculative Sampling Charlie Chen et al. (2023)
  推理
• FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning Tri Dao (2023)
  推理