SeerAttention/SeerAttention-DeepSeek-R1-Distill-Qwen-14B-AttnGates

SeerAttention-DeepSeek-R1-Distill-Qwen-14B-AttnGates

This repo only contains the AttnGates' weights for deepseek-ai/DeepSeek-R1-Distill-Qwen-14B.

SeerAttention introduces learnable AttnGate modules to accelerate the computationally intensive prefill stage of long-context large language models (LLMs) via dynamic block-level sparsity. The AttnGates are trained in a parameter-efficient self-distillation framework, where they learn to mimic the 2D max-pooled attention patterns of the original frozen model, preserving its integrity while avoiding costly retraining. During inference, these gates generate block-sparse binary masks by applying threshold/TopK to their learned soft scores, enabling efficient computation through a custom block-sparse FlashAttention kernel.

Original Github Repo https://github.com/microsoft/SeerAttention.