ExternalVendorCode/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2024-12-19 20:57:54 +00:00
Code Issues Actions 13 Packages Projects Releases Wiki Activity

models(gallery): add llama-3-perky-pat-instruct-8b (#2740)

Browse Source 
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2024-07-07 15:10:37 +02:00 committed by GitHub
parent 748e4cb6b1
commit 3a88299cfe
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 24 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
gallery/index.yaml
View File

@ -2435,6 +2435,30 @@
- filename: Llama3-8B-Instruct-Replete-Adapted-Q4_K_M.gguf
sha256: 9e9a142f6fb5fc812b17bfc30230582ae50ac22b93dea696b6887cde815c1cb4
uri: huggingface://bartowski/Llama3-8B-Instruct-Replete-Adapted-GGUF/Llama3-8B-Instruct-Replete-Adapted-Q4_K_M.gguf
- !!merge <<: *llama3
name: "llama-3-perky-pat-instruct-8b"
urls:
- https://huggingface.co/grimjim/Llama-3-Perky-Pat-Instruct-8B
- https://huggingface.co/bartowski/Llama-3-Perky-Pat-Instruct-8B-GGUF
description: |
we explore negative weight merger, and propose Orthogonalized Vector Adaptation, or OVA.
This is a merge of pre-trained language models created using mergekit.
"One must imagine Sisyphys happy."
Task arithmetic was used to invert the intervention vector that was applied in MopeyMule, via application of negative weight -1.0. The combination of model weights (Instruct - MopeyMule) comprises an Orthogonalized Vector Adaptation that can subsequently be applied to the base Instruct model, and could in principle be applied to other models derived from fine-tuning the Instruct model.
This model is meant to continue exploration of behavioral changes that can be achieved via orthogonalized steering. The result appears to be more enthusiastic and lengthy responses in chat, though it is also clear that the merged model has some unhealed damage.
Built with Meta Llama 3.
overrides:
parameters:
model: Llama-3-Perky-Pat-Instruct-8B-Q4_K_M.gguf
files:
- filename: Llama-3-Perky-Pat-Instruct-8B-Q4_K_M.gguf
sha256: b0eae5d9d58a7101a30693c267097a90f4a005c81fda801b40ab2c25e788a93e
uri: huggingface://bartowski/Llama-3-Perky-Pat-Instruct-8B-GGUF/Llama-3-Perky-Pat-Instruct-8B-Q4_K_M.gguf
- &chatml
### ChatML
url: "github:mudler/LocalAI/gallery/chatml.yaml@master"