DeepSeek-V3.2
DeepSeekTherefore, we implement DSA based on the MQA (Shazeer, 2019) mode of MLA, where each latent vector (the key-value entry of MLA) will be shared across all query heads of the query token.
implementation detail · filed under model architecture
The MLA mode in which each latent key-value vector is shared across query heads.
The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See multi-head latent attention for the methods filed beside it.
2 quotes from 2 sources, one card per model, each opening with its strongest.
Therefore, we implement DSA based on the MQA (Shazeer, 2019) mode of MLA, where each latent vector (the key-value entry of MLA) will be shared across all query heads of the query token.