ExternalVendorCode/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2024-12-20 05:07:54 +00:00
Code Issues Actions 13 Packages Projects Releases Wiki Activity

models(gallery): add llama-3-llamilitary (#2711)

Browse Source 
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2024-07-04 17:57:38 +02:00 committed by GitHub
parent 9aec1b3a61
commit 23b926d43e
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 19 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
gallery/index.yaml
View File

@ -1405,6 +1405,25 @@
- filename: L3-8B-Everything-COT-Q4_K_M.gguf - filename: L3-8B-Everything-COT-Q4_K_M.gguf
sha256: b220b0e2f8fb1c8a491d10dbd054269ed078ee5e2e62dc9d2e3b97b06f52e987 sha256: b220b0e2f8fb1c8a491d10dbd054269ed078ee5e2e62dc9d2e3b97b06f52e987
uri: huggingface://bartowski/L3-8B-Everything-COT-GGUF/L3-8B-Everything-COT-Q4_K_M.gguf uri: huggingface://bartowski/L3-8B-Everything-COT-GGUF/L3-8B-Everything-COT-Q4_K_M.gguf
- !!merge <<: *llama3
name: "llama-3-llamilitary"
urls:
- https://huggingface.co/Heralax/llama-3-llamilitary
- https://huggingface.co/mudler/llama-3-llamilitary-Q4_K_M-GGUF
icon: https://cdn-uploads.huggingface.co/production/uploads/64825ebceb4befee377cf8ac/ea2C9laq24V6OuxwhzJZS.png
description: |
This is a model trained on [instruct data generated from old historical war books] as well as on the books themselves, with the goal of creating a joke LLM knowledgeable about the (long gone) kind of warfare involving muskets, cavalry, and cannon.
This model can provide good answers, but it turned out to be pretty fragile during conversation for some reason: open-ended questions can make it spout nonsense. Asking facts is more reliable but not guaranteed to work.
The basic guide to getting good answers is: be specific with your questions. Use specific terms and define a concrete scenario, if you can, otherwise the LLM will often hallucinate the rest. I think the issue was that I did not train with a large enough system prompt: not enough latent space is being activated by default. (I'll try to correct this in future runs).
overrides:
parameters:
model: llama-3-llamilitary-q4_k_m.gguf
files:
- filename: llama-3-llamilitary-q4_k_m.gguf
sha256: f3684f2f0845f9aead884fa9a52ea67bed53856ebeedef1620ca863aba57e458
uri: huggingface://mudler/llama-3-llamilitary-Q4_K_M-GGUF/llama-3-llamilitary-q4_k_m.gguf
- &dolphin - &dolphin
name: "dolphin-2.9-llama3-8b" name: "dolphin-2.9-llama3-8b"
url: "github:mudler/LocalAI/gallery/hermes-2-pro-mistral.yaml@master" url: "github:mudler/LocalAI/gallery/hermes-2-pro-mistral.yaml@master"