Skip to content

Small correction of MLA acronym in README.md #5

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 8, 2025
Merged

Small correction of MLA acronym in README.md #5

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion deepseek_r1_jax/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -93,7 +93,7 @@ HBM leading to >100% utilization.
### Optimization Decisions

Deepseek is a unique model in that it (i) uses a unique form of attention,
Multi-Head Attention (MLA) and (ii) uses an MoE layer with a large number of
Multi-head Latent Attention (MLA) and (ii) uses an MoE layer with a large number of
small experts. This presents some challenges in optimizing the model for
TPUs and GPUs to maximize either compute (in training) or memory-bandwidth use
in inference.
Expand Down