Skip to content
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

Jump to bottom

Fix QKV weight sharding for gemma #222

Merged
merged 1 commit into from
Feb 27, 2024
Merged

Fix QKV weight sharding for gemma #222

merged 1 commit into from
Feb 27, 2024

Conversation

masahi
Copy link
Member

@masahi masahi commented Feb 26, 2024 •
edited
Loading

This fixes multi-gpu inference for gemma. The sharding function was assuming head_dim = model_config.hidden_size // model_config.num_attention_heads, which is incorrect for gemma.

@sunggg @Lunderberg

@masahi masahi merged commit 0dfb756 into octoml:batch-serving Feb 27, 2024
1 check passed
@masahi masahi deleted the gemma-followup branch February 27, 2024 07:27
Lunderberg pushed a commit to Lunderberg/mlc-llm that referenced this pull request Feb 27, 2024
@spectrometerHBH
Remove std::filesystem::canonicalize (octoml#222)
f1dcc7f
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@masahi