llama.cpp
Allow pooled embeddings on any model
#7477
Merged
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
6
Changes
View On
GitHub
Allow pooled embeddings on any model
#7477
ggerganov
merged 6 commits into
ggml-org:master
from
iamlemec:append-pooling
github-actions
added
examples
compilade
commented on 2024-05-22
compilade
added
enhancement
compilade
added
embeddings
compilade
added
Review Complexity : Medium
compilade
commented on 2024-06-01
ggerganov
commented on 2024-06-04
ggerganov
approved these changes on 2024-06-07
ggerganov
requested a review
from
compilade
1 year ago
create append_pooling operation; allow to specify attention_type; add…
01057149
find result_norm/result_embd tensors properly; update output allocati…
1756c4b5
only use embd output for pooling_type NONE
7c37ae9d
get rid of old causal_attn accessor
d4e6972f
take out attention_type; add in llama_set_embeddings
8093253b
iamlemec
force pushed
to
8093253b
1 year ago
compilade
commented on 2024-06-14
bypass logits when doing non-NONE pooling
5cc7b453
compilade
approved these changes on 2024-06-20
ngxson
commented on 2024-06-20
ggerganov
merged
80ea089d
into master
1 year ago
Login to write a write a comment.
Login via GitHub
Reviewers
ggerganov
compilade
ngxson
Assignees
No one assigned
Labels
enhancement
examples
embeddings
Review Complexity : Medium
Milestone
No milestone
Login to write a write a comment.
Login via GitHub