DavidAU commited on
Commit
11d8a8c
·
verified ·
1 Parent(s): 1222abc

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +85 -0
README.md ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: transformers
4
+ language:
5
+ - en
6
+ - fr
7
+ - zh
8
+ - de
9
+ tags:
10
+ - creative
11
+ - creative writing
12
+ - fiction writing
13
+ - plot generation
14
+ - sub-plot generation
15
+ - fiction writing
16
+ - story generation
17
+ - scene continue
18
+ - storytelling
19
+ - fiction story
20
+ - science fiction
21
+ - romance
22
+ - all genres
23
+ - story
24
+ - writing
25
+ - vivid prose
26
+ - vivid writing
27
+ - moe
28
+ - mixture of experts
29
+ - 128 experts
30
+ - 8 active experts
31
+ - fiction
32
+ - roleplaying
33
+ - bfloat16
34
+ - rp
35
+ - qwen3
36
+ - horror
37
+ - finetune
38
+ - thinking
39
+ - reasoning
40
+ - qwen3_moe
41
+ - merge
42
+ - uncensored
43
+ - abliterated
44
+ - not-for-all-audiences
45
+ base_model:
46
+ - huihui-ai/Qwen3-30B-A3B-abliterated
47
+ pipeline_tag: text-generation
48
+ ---
49
+
50
+ ( Uploading, Quants pending, model card updates pending, examples to be added )
51
+
52
+ <h2>Qwen3-33B-A3B-Stranger-Thoughts-Abliterated-Uncensored</h2>
53
+
54
+ This repo contains the full precision source code, in "safe tensors" format to generate GGUFs, GPTQ, EXL2, AWQ, HQQ and other formats.
55
+ The source code can also be used directly.
56
+
57
+ ABOUT:
58
+
59
+ A stranger, yet radically different version of "Qwen/Qwen3-30B-A3B", abliterated by "huihui-ai" , with 4 added layers expanding
60
+ the model to 33B total parameters.
61
+
62
+ The goal: slightly alter the model, to address some odd creative thinking and output choices AND de-censor it.
63
+
64
+ Please note that the modifications affect the entire model operation; roughly I adjusted the model to think a little "deeper"
65
+ and "ponder" a bit - but this is a very rough description.
66
+
67
+ I also ran reasoning tests (non-creative) to ensure model was not damaged and roughly matched original model performance.
68
+
69
+ That being said, reasoning and output generation will be altered regardless of your use case(s).
70
+
71
+ USAGE:
72
+
73
+ Please refer to this model card for Abliterated model specific prompting, specific usage, suggested settings, changing ACTIVE EXPERTS, templates, settings and the like:
74
+
75
+ https://huggingface.co/DavidAU/Qwen3-18B-A3B-Stranger-Thoughts-Abliterated-Uncensored-GGUF
76
+
77
+ and
78
+
79
+ https://huggingface.co/DavidAU/Qwen3-33B-A3B-Stranger-Thoughts-GGUF
80
+
81
+ ---
82
+
83
+ EXAMPLES Coming soon...
84
+
85
+ ---