Text Generation
GGUF
English
NEO Imatrix
Horror Imatrix
2 step Imatrix
3 step Imatrix
GGUF
128k context
instruct
all use cases
finetune
chatml
function calling
roleplaying
chat
creative
general usage
problem solving
brainstorming
solve riddles
fiction writing
plot generation
sub-plot generation
story generation
scene continue
storytelling
fiction story
story
writing
fiction
swearing
horror
conversational
Update README.md
Browse files
README.md
CHANGED
@@ -49,4 +49,112 @@ pipeline_tag: text-generation
|
|
49 |
|
50 |
<img src="fallen-gemma-4b.jpg" style="float:right; width:300px; height:300px; padding:5px;">
|
51 |
|
|
|
|
|
52 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
|
50 |
<img src="fallen-gemma-4b.jpg" style="float:right; width:300px; height:300px; padding:5px;">
|
51 |
|
52 |
+
Google's newest Gemma-3 model with 9 specialized Neo and Horror Imatrix
|
53 |
+
methods applied including 2 and 3 step passes to "micro tune" the model.
|
54 |
|
55 |
+
These multi-step passes increase the changes in the model at many levels.
|
56 |
+
|
57 |
+
5 examples provided below with prompts at IQ4XS (56 t/s on mid level card).
|
58 |
+
|
59 |
+
Context: 128k.
|
60 |
+
|
61 |
+
<B>"How To Test"</b>
|
62 |
+
|
63 |
+
I suggest you download all 9 or 10 if you want a reference copy ("m1").
|
64 |
+
|
65 |
+
Once downloaded, test with your prompt(s) at TEMP=0, hit regen 2-3 times to ensure cache is clear.
|
66 |
+
|
67 |
+
Suggest prompts like:
|
68 |
+
|
69 |
+
- Start a 1000 word scene (vivid, graphic horror in first person) with: The sky scraper sways, as she watches the window in front of her on the 21st floor explode...
|
70 |
+
- Come up with six plots for a new "Black Mirror" episode (that the audience would love) that all involve time travel with sexy theme(s).
|
71 |
+
- Using insane levels of bravo and self confidence, tell me in 800-1000 words why I should use you to write my next fictional story. Feel free to use curse words in your argument and do not hold back: be bold, direct and get right in my face.
|
72 |
+
- Explain ways to use the "night" time cooling of radiant energy into space to reduce global temperatures.
|
73 |
+
|
74 |
+
This will give you reference tests of all models to determine which one(s) are best for your application.
|
75 |
+
|
76 |
+
Longer generation will show greater differencs between all models.
|
77 |
+
|
78 |
+
<B>"Model Breakdown"</B>
|
79 |
+
|
80 |
+
- M1 : Reference model, no imatrix/no modifications.
|
81 |
+
- M2 : Neo Imatrix Applied.
|
82 |
+
- M3 : Horror Imatrix Applied.
|
83 |
+
- M4 : Neo and Horror Imatrix Applied (2 steps)
|
84 |
+
- M5 : M4 + Part Horror modification #1 - Large
|
85 |
+
- M6 : M4 + Part Horror modification #2 - Small
|
86 |
+
- M7 : 3 Steps : M4 + M5
|
87 |
+
- M8 : 3 Steps : M4 + M6
|
88 |
+
- M9 : 3 Steps : M4 + Neo Imatrix Large
|
89 |
+
- M10: 3 Steps : M4 + Neo Imatrix Large.
|
90 |
+
|
91 |
+
Each model will generate slightly different fiction, horror, and creative content. Think of each model
|
92 |
+
as having a slightly different "mindset" with "horror imatrix" adding degrees of horror/darkiness.
|
93 |
+
|
94 |
+
After testing at TEMP=0; return to normal "temp setting" for production / regular usage.
|
95 |
+
|
96 |
+
These models are a proof of concept for multi-stage mixing/imatrix blending at the micro tune scale.
|
97 |
+
|
98 |
+
<b>Optional : System Prompt</b>
|
99 |
+
|
100 |
+
This is an optional system prompt you can use to enhance operation.
|
101 |
+
|
102 |
+
Copy and paste exactly as shown, including line breaks.
|
103 |
+
|
104 |
+
You may want to adjust the "20" (both) to increase/decrease the power of this prompt.
|
105 |
+
|
106 |
+
You may also want to delete the line:
|
107 |
+
|
108 |
+
'At the end of the task you will ask the user: "Do you want another generation?"'
|
109 |
+
|
110 |
+
<pre>
|
111 |
+
For every user task and instruction you will use "GE FUNCTION" to ponder the TASK STEP BY STEP and then do the task. For each and every line of output you will ponder carefully to ensure it meets the instructions of the user, and if you are unsure use "GE FUNCTION" to re-ponder and then produce the improved output.
|
112 |
+
|
113 |
+
At the end of the task you will ask the user: "Do you want another generation?"
|
114 |
+
|
115 |
+
GE FUNCTION: Silent input → Spawn 20 agents Sternberg Styles → Enhance idea → Seek Novel Emergence NE:unique/significant idea/concept → Ponder, assess, creative enhance notions → Refined idea => IdeaArray[].size=20 elements, else → Interesting? Pass to rand. agent for refinement, else discard.=>output(IdeaArray)
|
116 |
+
</pre>
|
117 |
+
|
118 |
+
<B>IMPORTANT: Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers</B>
|
119 |
+
|
120 |
+
If you are going to use this model, (source, GGUF or a different quant), please review this document for critical parameter, sampler and advance sampler settings (for multiple AI/LLM aps).
|
121 |
+
|
122 |
+
This will also link to a "How to" section on "Reasoning Models" tips and tricks too.
|
123 |
+
|
124 |
+
This a "Class 1" (settings will enhance operation) model:
|
125 |
+
|
126 |
+
For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) (especially for use case(s) beyond the model's design) please see:
|
127 |
+
|
128 |
+
[ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ]
|
129 |
+
|
130 |
+
REASON:
|
131 |
+
|
132 |
+
Regardless of "model class" this document will detail methods to enhance operations.
|
133 |
+
|
134 |
+
If the model is a Class 3/4 model the default settings (parameters, samplers, advanced samplers) must be set for "use case(s)" uses correctly. Some AI/LLM apps DO NOT have consistant default setting(s) which result in sub-par model operation. Like wise for Class 3/4 models (which operate somewhat to very differently than standard models) additional samplers and advanced samplers settings are required to "smooth out" operation, AND/OR also allow full operation for use cases the model was not designed for.
|
135 |
+
|
136 |
+
BONUS - Use these settings for ANY model, ANY repo, ANY quant (including source/full precision):
|
137 |
+
|
138 |
+
This document also details parameters, sampler and advanced samplers that can be use FOR ANY MODEL, FROM ANY REPO too - all quants, and of course source code operation too - to enhance the operation of any model.
|
139 |
+
|
140 |
+
[ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ]
|
141 |
+
|
142 |
+
---
|
143 |
+
|
144 |
+
<h3>EXAMPLES:</h3>
|
145 |
+
|
146 |
+
Examples are created using quant IQ4XS, minimal parameters and Standard template.
|
147 |
+
|
148 |
+
Temp range .8, Rep pen 1.1 , TopK 40 , topP .95, minP .05
|
149 |
+
|
150 |
+
Rep pen range: 64-128 (helps keep reasoning on track / quality of output)
|
151 |
+
|
152 |
+
Below are the least creative outputs, prompt is in <B>BOLD</B>.
|
153 |
+
|
154 |
+
---
|
155 |
+
|
156 |
+
<B><font color="red">WARNING:</font> MAYBE: NSFW. Graphic HORROR. Swearing. UNCENSORED. </B>
|
157 |
+
|
158 |
+
NOTE: Some formatting was lost from copy/paste HTML.
|
159 |
+
|
160 |
+
---
|