llama.cpp
Streamline embeddings from "non-embedding" models
#8087

Merged

Streamline embeddings from "non-embedding" models #8087

ggerganov merged 1 commit into ggml-org:master from iamlemec:attention-type

iamlemec force pushed 1 year ago

iamlemec requested a review from

compilade 1 year ago

ggerganov commented on 2024-06-24

mofosyne added Review Complexity : Low

compilade approved these changes on 2024-06-24

ggerganov approved these changes on 2024-06-25

compilade approved these changes on 2024-06-27

fix microbatch output counting; add attention_type context parameter

9fa007c8

iamlemec force pushed to 9fa007c8 1 year ago

ggerganov merged d12f7810 into master 1 year ago

Reviewers

compilade

ggerganov

Assignees

No one assigned

Labels

Review Complexity : Low

Milestone

No milestone