Pinned Loading

1. Awesome-Attention-Sink Awesome-Attention-Sink Public

   🚀 First survey on Attention Sink in Transformers — 200+ papers on utilization, interpretation, and mitigation.

   143 9

2. OScaR-KV-Quant OScaR-KV-Quant Public

   🏆 OScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyond — redefining the accuracy-efficiency Pareto front for X-LLMs KV quantization.

   C++ 142 15

3. rattlesnakey/Awesome-Actionable-MI-Survey rattlesnakey/Awesome-Actionable-MI-Survey Public

   The Github repo for our survey paper: "Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models"

   160 13

4. meituan-longcat/SGLang-FluentLLM meituan-longcat/SGLang-FluentLLM Public

   Python 119 16

5. meituan-longcat/LongCat-Flash-Chat meituan-longcat/LongCat-Flash-Chat Public

   1.4k 76

6. StartLuxLabs/Massive-Activations-HLA StartLuxLabs/Massive-Activations-HLA Public

   ⚡ Official code for “Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus”.

   Python 19 3