Native Sparse Attention PyTorch Implementaion (non-official)
Implementation of Native Sparse Attention(NSA) from Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention in PyTorch.
Thank you, DeepSeek-AI.
| Name | Name | Last commit date | ||
|---|---|---|---|---|
Native Sparse Attention PyTorch Implementaion (non-official)
Implementation of Native Sparse Attention(NSA) from Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention in PyTorch.
Thank you, DeepSeek-AI.
