Revival of my old DJZ Cyber Society Collection, <a target="_blank" rel="ugc" href="https://www.amazon.co.uk/Cyber-Society-1-Freddy-King/dp/B0B8VLGZMB">here is my book published in 2022</a> <a target="_blank" rel="ugc" href="https://www.youtube.com/live/O3lqADlLQuc">Livestream explanation of this project</a> V0 = fast trained, OG dataV00 = full trained, OG dataV1 = fast trained, Mutation StackV11 = full trained, Mutation Stack V2 = fast trained, Remaster Stack (200 character caption limit)V22 = full trained, Remaster Stack V3 = fast trained, Remaster Stack (200 token caption limit)V33 = full trained, Remaster Stack <a target="_blank" rel="ugc" href="https://civarchive.com/posts/7497496">Version comparison Gallery</a> To build the next generation of datasets, we let them decide, through a best of 1000 images, 8 way decisive battle. This puts all previous versions to the test. all 8 previous lora were tested with all 4 prompt version lists. The best 500 images were then curated down to ~300 and sent for recaptioning. The resulting dataset is version 4. V4 = fast trained, includes V0, winning images from the battle V44 = over 10,000 steps due to size of dataset. (shelved for now) V55 = Full train with culled V4 dataset. V00 &amp; Knight subset data removed. By making use of V4, in combinations with many prompt lists i wrote and by injecting combinations of 3 to 5 of my pretrained style in lora stacks - we generate over 2500 images. This workpile was curated down to 250 images, not for quality as all the images are excellent, but for aesthetic qualities that can teach a new and unique style which defines the look I am developing with this process. V6 = Fast Train of the 2500 images generated with V5 + Stacks, curated down to 250, then another 60 images pulled to prevent an anime bias in the data set. V66 = Full Train of the V6 dataset. Interestingly, common concepts will be diluted by my developed style and barely represent the strong images defined in the base model.example: "Batman" added to the prompt while properly invoking this trained style, will result in an images that is heavily styled in favor of my work and only the strongest features will remain. <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/9c5043f7-a1f0-4fc5-aa71-3b1edd419201/width=525/9c5043f7-a1f0-4fc5-aa71-3b1edd419201.jpeg" />as shown above, only the batman chest logo and bat ears are still recognisable. This shows that our style training is successfully being taught. Workflows and a video <s>will arrive</s> arrived, explaining what dataset remix mutation is and how you can use it.PROMPT REPORT: (V66)<h1 id="cybersociety-image-model-analysis-npii2z24v">CyberSociety Image Model Analysis</h1><h2 id="overview-y58zkgb5u">Overview</h2>This document analyzes the prompt structure and content patterns of the "CyberSociety" image generation model based on its training dataset. The model appears specialized in generating futuristic, sci-fi imagery with consistent aesthetic elements and themes.<h2 id="prompt-structure-6tzl92nin">Prompt Structure</h2>Each prompt in the dataset follows a consistent pattern:<pre><code>cybersociety style, , [detailed description], [comma-separated tags]
</code></pre>Key components:<ul><li>Style trigger: "cybersociety style" (appears in 100% of prompts)</li><li>Empty field: Consistently contains a blank between two commas</li><li>Detailed description: Elaborate paragraph describing the image content</li><li>Tags: Short, comma-separated keywords denoting key visual elements</li></ul><h2 id="trigger-words-and-associations-tic0npu5a">Trigger Words and Associations</h2>The primary trigger phrase "cybersociety style" is consistently associated with a distinct aesthetic featuring:<ul><li>Futuristic sci-fi imagery</li><li>Rich color palettes dominated by orange, blue, and red tones</li><li>Mechanical and technological elements</li><li>Dramatic atmospheric effects (clouds, lighting)</li><li>Detailed armor and equipment designs</li><li>Post-apocalyptic or dystopian settings</li></ul><h2 id="thematic-elements-and-common-subjects-jv4n7ikm6">Thematic Elements and Common Subjects</h2><h3 id="character-types-vmb4g1lge">Character Types</h3><ol><li>Astronauts/Space Explorers (extremely common)<ul><li>Always wearing elaborate helmets with visors</li><li>Suits adorned with various gadgets, tools, and mechanical components</li><li>Often carrying weapons or exploration equipment</li></ul></li><li>Armored Warriors/Soldiers<ul><li>Advanced combat gear</li><li>Futuristic weapons (both melee and ranged)</li><li>Helmets with visors or glowing elements</li></ul></li><li>Robots/Cyborgs<ul><li>Mechanical humanoids</li><li>Often with glowing eyes (particularly red)</li><li>Intricate mechanical components</li></ul></li></ol><h3 id="environment-elements-llfssmiil">Environment Elements</h3><ol><li>Skyscape Features<ul><li>Dramatic clouds (nearly universal)</li><li>Multiple moons/planets visible</li><li>Orange/red atmospheric tones</li><li>Birds in flight (surprisingly common)</li></ul></li><li>Architectural Elements<ul><li>Floating structures/islands</li><li>Domed buildings</li><li>Towers with spires</li><li>Industrial pipes and machinery</li><li>Clock towers (recurring motif)</li><li>Space stations</li></ul></li><li>Vehicles/Spacecraft<ul><li>Flying machines</li><li>Spherical crafts</li><li>Steampunk-inspired airships</li><li>Traditional rockets</li></ul></li></ol><h2 id="common-phrases-and-language-patterns-ouifah8zm">Common Phrases and Language Patterns</h2>The descriptions follow remarkably consistent language patterns:<ol><li>Opening phrases:<ul><li>"The image depicts a futuristic scene with..." (extremely common)</li><li>"The image depicts a character in..."</li><li>"The image depicts two astronauts in..."</li></ul></li><li>Character descriptions:<ul><li>"clad in an elaborate/intricate/detailed space suit/armor"</li><li>"wearing a helmet with a visor"</li><li>"adorned with various gadgets and tools"</li><li>"holding a weapon/sword/gun"</li></ul></li><li>Background descriptions:<ul><li>"set against a backdrop of..."</li><li>"dramatic sky filled with clouds"</li><li>"sky is a gradient of warm colors"</li><li>"in the background, there is/are..."</li></ul></li><li>Color descriptors:<ul><li>"rich with oranges, blues, and metallic tones"</li><li>"warm tones dominating"</li><li>"clouds that have an orange and red hue"</li><li>"sense of depth and complexity"</li></ul></li><li>Atmospheric qualifiers:<ul><li>"suggesting either dawn or dusk"</li><li>"creating a sense of tension"</li><li>"adding to the otherworldly atmosphere"</li><li>"evoking a sense of advanced technology"</li></ul></li></ol><h2 id="tag-analysis-hdhaqwee1">Tag Analysis</h2>The most frequent tags in order of prominence:<ol><li>Character elements:<ul><li>helmet (appears in ~80% of prompts)</li><li>astronaut/space</li><li>person</li><li>armor</li><li>cyborg/robot</li></ul></li><li>Environmental elements:<ul><li>clouds (appears in ~75% of prompts)</li><li>sky</li><li>futuristic</li><li>buildings/cityscape</li><li>moon/planet</li></ul></li><li>Gear/equipment:<ul><li>weapon/gun/sword</li><li>suit</li><li>spaceship/spacecraft</li></ul></li><li>Color tags:<ul><li>orange</li><li>blue</li><li>red</li></ul></li><li>Miscellaneous recurring elements:<ul><li>bird (surprisingly common)</li><li>tower</li><li>smoke</li><li>cape</li></ul></li></ol><h2 id="model-capabilities-range-fhvlsm9m0">Model Capabilities Range</h2>The dataset indicates the model is optimized for:<ol><li>Character-focused compositions<ul><li>Single figure prominently featured (most common)</li><li>Two figures interacting or posed together</li><li>Small groups (3-5 figures) less common</li><li>Both male and female characters represented</li></ul></li><li>Environmental scope<ul><li>Close/medium views with character focus (most common)</li><li>Wide landscape shots with small figures and grand vistas</li><li>Industrial/mechanical interiors</li><li>Space stations/spacecraft interiors</li><li>Planetary surfaces (barren/rocky most common)</li></ul></li><li>Lighting conditions<ul><li>Dramatic sunset/sunrise lighting (extremely common)</li><li>High contrast scenes</li><li>Silhouette effects</li><li>Glowing light sources (moons, planets, mechanical elements)</li></ul></li><li>Stylistic range<ul><li>Primarily science fiction aesthetic</li><li>Heavy steampunk influences</li><li>Post-apocalyptic elements</li><li>Military/combat scenarios</li><li>Exploration themes</li></ul></li></ol><h2 id="example-prompt-categories-cqqx0y6dc">Example Prompt Categories</h2><h3 id="character-focused-z03ka5jpi">Character-Focused</h3><pre><code>cybersociety style, , The image depicts a futuristic astronaut in a detailed, ornate space suit with various gadgets and tools attached. The suit is adorned with various gadgets, dials, and pipes, giving it an intricate and mechanical appearance. The helmet has a visor and is decorated with intricate patterns. In the background, there's a large mechanical object resembling a satellite or space station, with a prominent circular window and complex machinery. The color palette consists of deep blues, oranges, and grays, creating a moody and atmospheric scene., person, astronaut, space, clouds, sky, helmet, suit, orange, blue, clock
</code></pre><h3 id="landscapeenvironment-focused-fjln029hv">Landscape/Environment-Focused</h3><pre><code>cybersociety style, , The image depicts a futuristic, dystopian cityscape with a large, spherical moon dominating the sky. The architecture is a mix of industrial and ancient styles, with domed structures and spires that have an otherworldly appearance. In the foreground, there are several figures wearing futuristic armor, some holding weapons, suggesting a scene of conflict or defense. The color palette is dominated by warm tones, with oranges and reds creating a sense of tension and urgency. The overall atmosphere is one of a post-apocalyptic world, possibly on another planet, with advanced technology and a desolate environment., space, planet, moon, buildings, people, weapons, helmet, sky
</code></pre><h3 id="actioncombat-focused-l3ttdmi4s">Action/Combat-Focused</h3><pre><code>cybersociety style, , The image depicts a futuristic warrior standing amidst a dramatic, cloud-filled sky. The warrior is clad in an intricate armor suit with a helmet featuring a visor and a red glow around the eyes, suggesting advanced technology. They hold a staff or weapon with a glowing tip, indicating readiness for combat. Surrounding the warrior are floating structures and mechanical devices, hinting at a high-tech, possibly dystopian environment. The color palette is rich with deep blues, oranges, and reds, creating a sense of tension and otherworldliness., cyborg, futuristic, armor, helmet, red eyes, clouds, sky, floating structures, spacecraft, dystopian
</code></pre><h3 id="vehicletechnology-focused-qxunpqv7p">Vehicle/Technology-Focused</h3><pre><code>cybersociety style, , The image depicts a futuristic scene with a central figure standing atop a spherical structure, possibly a space station or a futuristic city. The character is clad in detailed armor and wields two swords, suggesting a warrior or guardian role. In the background, there are multiple spacecraft of varying sizes, indicating advanced technology and space exploration. The sky is filled with clouds that have an orange and pink hue, giving the impression of either dawn or dusk. The architecture below is dark and industrial, with tall buildings and structures that resemble a dystopian or post-apocalyptic setting. The overall atmosphere is one of a sci-fi world where human civilization has evolved into a technologically advanced society with a focus on space travel and defense., cyborg, futuristic cityscape, clouds, spaceship, armor, helmet, sword
</code></pre><h2 id="conclusion-wy80ada38">Conclusion</h2>The CyberSociety image model demonstrates a highly specialized focus on futuristic, sci-fi imagery with remarkably consistent visual elements. The model appears to be trained on a dataset with strong internal coherence, featuring recurring characters (astronauts, warriors), environments (dramatic skies, mechanical structures), and color palettes (orange/blue/red dominance).The prompt structure utilizes the "cybersociety style" trigger to activate a specific aesthetic, followed by detailed descriptions that establish scene composition, character elements, environmental factors, and mood/atmosphere. This is complemented by tag lists that reinforce key visual elements.The model demonstrates particular strength in rendering helmeted figures in elaborate technological gear against dramatic skyscapes, with a consistent stylistic approach that blends sci-fi, steampunk, and post-apocalyptic aesthetics into a cohesive visual language.

DJZ CyberSociety

blacksun-V1-captions.zip

cybersocietyV11-e9.safetensors

using the dataset re-captioning process shown here: Black Sun V0 = Florence derived captions Black Sun V1 = UfromGen2Qwen derived captions retrained an old 2.1 768 dataset, using flux to remix and recaption for ClipL/T5

Files with SHA256: e927be1d5066737bd6ced39394f052414ca108370d8347b6abd8d629929b7291