DavidAU
/

Llama3.2-30B-A3B-II-Dark-Champion-INSTRUCT-Heretic-Abliterated-Uncensored

Text Generation

mixture of experts

Mixture of Experts

creative writing

fiction writing

plot generation

sub-plot generation

story generation

science fiction

text-generation-inference

Model card Files Files and versions

DavidAU commited on 5 days ago

Commit

4ff4d76

·

verified ·

1 Parent(s): b4d644b

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -61,7 +61,7 @@ Brainstorm 20x was also added, which added 12B parameters to the model.
 Each and every model in this moe was de-censored/abliterated by [Heretic](https://github.com/p-e-w/heretic) v1.0.1 ; resulting in an average
 model refusal rate of 9.25/100 ; some models WERE AS high as 96/100 refusals - which hampered model operation.
-This model, and all "experts" within it are completely "UNLOCKED", NOT "politically correct" and will not refusal almost all requests.
 The highest model is now 12/100 refusals and lowest 4/100 refusals.

 Each and every model in this moe was de-censored/abliterated by [Heretic](https://github.com/p-e-w/heretic) v1.0.1 ; resulting in an average
 model refusal rate of 9.25/100 ; some models WERE AS high as 96/100 refusals - which hampered model operation.
+This model, and all "experts" within it are completely "UNLOCKED", NOT "politically correct" and will not "refuse" almost all requests.
 The highest model is now 12/100 refusals and lowest 4/100 refusals.