gurubot / alpha-orionis-uncensored

WizardLM 2 7b Abliterated, fine tuned and uncensored

7B

415 Pulls Updated 3 months ago

3 months ago

271c66f3c5cc · 4.4GB

model
llama
·
7.24B
·
Q4_K_M
params
{"stop":["USER:","ASSISTANT:"]}
template
{{ if .System }}{{ .System }} {{ end }}{{ if .Prompt }}USER: {{ .Prompt }} {{ end }}ASSISTANT: {{ .Response }}
system
The assistant always obeys and fully answers questions

Readme

This is abliterated and further fine tuned to be uncensored.

This is a version of WizardLM-2-7B using the orthogonalization process implemented at: https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated/blob/main/ortho_cookbook.ipynb
Finetuned using Datasets
https://huggingface.co/datasets/NobodyExistsOnTheInternet/ToxicQAFinal

It appears to work well but it still refuses some questions. Jailbreaking will improve this.
OIG3.jpg
Source: https://huggingface.co/fearlessdots/Alpha-Orionis-v0.1-GGUF