✳flâneur — a map of the web's best reading

On the small model, the actual GPU memory usage of Mamba2 is much higher than that of Mamba1. · Issue #439 · state-spaces/mamba

github.com · 323 words · saved by 1 readers

The parameters of the Mamba2 model are d_state=32, d_conv=4, expand=2, and head_dim=32 (using "nn. Conv1d" with padding method, without the constraint of d_model/head_dim%8==0). Mamba1 maintains the same parameters except for the absence of head_dim. Although the inference speed of Mamba2 has almost doubled compared to Mamba1, the actual memory usage has increased from 4.82G to 7.55G (in my task). I would like to ask if this is due to the basic computational load of Mamba2's semi separation matrix, which poses a disadvantage in small-scale models? I see in your paper that on larger scale models, the actual memory usage of Mamba2 is lower. The text was updated successfully, but these errors were encountered: nn.Conv1d is probably not great for memory usage. You should try to use causal_conv1d. Sorry, something went wrong. nn.Conv1d is probably not great for memory usage. You should try to use causal_conv1d. Okay, thank you for your rep

On the small model, the actual GPU memory usage of Mamba2 is much higher than that of Mamba1. · Issue #439 · state-spaces/mamba · GitHub Skip to content You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert {{ message }} Uh oh! There was an error while loading. Please reload this page . state-spaces / mamba Public Notifications You must be signed in to change notification settings Fork 1.8k Star 18.6k On the s

Explore this link on the map →

related reading