bird-of-paradise
/

deepseek-mla

Text Generation

attention-mechanism

efficient-attention

Model card Files Files and versions Community

deepseek-mla / assets

2 contributors

History: 1 commit

Yan Wei

Initial commit: DeepSeek Multi-Latent Attention implementation

550eb56 about 1 month ago

mla_architecture.png

346 kB

Initial commit: DeepSeek Multi-Latent Attention implementation about 1 month ago
mla_formulas.png

358 kB

Initial commit: DeepSeek Multi-Latent Attention implementation about 1 month ago