DavidAU/Qwen3-18B-A3B-Stranger-Thoughts-Abliterated-Uncensored-GGUF
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
David Belton
2025-06-17 01:29:26 +00:00
committed by system
parent 1d48ebff80
commit 4d97470bca
1 changed files with 27 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
README.md
View File

@@ -37,14 +37,20 @@ tags:
- thinking - thinking
- reasoning - reasoning
- qwen3_moe - qwen3_moe
- not-for-all-audiences
- uncensored
- abliterated
base_model: base_model:
- kalomaze/Qwen3-16B-A3B - kalomaze/Qwen3-16B-A3B
- huihui-ai/Moonlight-16B-A3B-Instruct-abliterated - huihui-ai/Qwen3-16B-A3B-abliterated
pipeline_tag: text-generation pipeline_tag: text-generation
--- ---
(quants uploading, examples to be added...) (quants uploading, examples to be added...)
<B><font color="red">WARNING:</font> NSFW. Vivid prose. INTENSE. Visceral Details. Violence. HORROR. GORE. Swearing. UNCENSORED... humor, romance, fun. </B>
<h2>Qwen3-18B-A3B-Stranger-Thoughts-Abliterated-Uncensored-GGUF</h2> <h2>Qwen3-18B-A3B-Stranger-Thoughts-Abliterated-Uncensored-GGUF</h2>
<img src="qwen-18b.jpg" style="float:right; width:300px; height:300px; padding:10px;"> <img src="qwen-18b.jpg" style="float:right; width:300px; height:300px; padding:10px;">
@@ -123,6 +129,26 @@ I found the best results during testing (stable thinking, generation, minimal is
- temp range .6 to 1.2 ; with outside max of 1.8 - temp range .6 to 1.2 ; with outside max of 1.8
- context of 8k min - context of 8k min
<B>IMPORTANT: Using an "abliterated" model VS "uncensored" model</B>
Usually when you a tell a model to generate horror, swear or x-rated content this is all you have to do to get said content type.
In the case of this model, it will not refuse your request, however it needs to be "pushed" a bit / directed a bit more in SOME CASES.
See examples #1 and #2 below show this.
Although this model will generated x-rated content too, likewise you need to tell it to use "slang" (and include the terms you want)
to get it generate the content correctly as the "expected" content level too.
Without these added directive(s), the content can be "bland" by comparison to an "uncensored model" or model trained on uncensored content.
Roughly, the model tries to generate the content but the "default" setting(s) are so "tame" it needs a push to generate at expected graphic,
cursing or explicit levels.
Even with minimal direction (ie, use these words to swear: x,y,z), this will be enough to push the model to generate the requested content in the ahh... expected format.
---
You may find that Qwen's default parameters/samplers also work better for your use case(s) too. You may find that Qwen's default parameters/samplers also work better for your use case(s) too.
Please refer to the orginal model repo for all default settings, templates, benchmarks, etc.: Please refer to the orginal model repo for all default settings, templates, benchmarks, etc.: