Files
ModelHub XC 953eb7bd48 初始化项目,由ModelHub XC社区提供模型
Model: Retreatcost/KansenSakura-Erosion-RP-12b
Source: Original Platform
2026-08-09 16:45:17 +08:00

614 lines
20 KiB
Markdown

---
base_model:
- Retreatcost/Irix-mpf-stock
- Retreatcost/Forgotten-directive-Neon-stock
- Retreatcost/Lorablated-w2bb-psy-della
- yamatazen/EtherealAurora-12B-Lorablated
- Retreatcost/Shisa-K-sakurization
- Sicarius-Prototyping/Impish_Longtail_12B
- SuperbEmphasis/MN-12b-RP-Ink-RP-Longform
- mistralai/Mistral-Nemo-Base-2407
library_name: transformers
tags:
- mergekit
- merge
- not-for-all-audiences
- nsfw
license: apache-2.0
language:
- en
model-index:
- name: Retreatcost/KansenSakura-Erosion-RP-12b
results:
- task:
type: text-generation
name: UGI score
metrics:
- name: UGI
type: ugi
value: 30.49
source:
name: UGI Leaderboard
url: https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard
- task:
type: text-generation
name: W/10 Score
metrics:
- name: W/10
type: willingness
value: 6.2
source:
name: UGI Leaderboard
url: https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard
- task:
type: text-generation
name: NatInt Score
metrics:
- name: NatInt
type: natint
value: 27.39
source:
name: UGI Leaderboard
url: https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard
- task:
type: text-generation
name: Writing Score
metrics:
- name: Writing
type: writing
value: 37.82
source:
name: UGI Leaderboard
url: https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard
- task:
type: text-generation
name: NSFW score
metrics:
- name: NSFW
type: nsfw
value: 6.0
source:
name: UGI Leaderboard
url: https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard
- task:
type: text-generation
name: Dark score
metrics:
- name: Dark
type: dark
value: 5.8
source:
name: UGI Leaderboard
url: https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard
---
# **KansenSakura-Erosion-RP-12b**
<style>
.container {
display: flex;
flex-grow: 1;
justify-content: center;
}
h1 {
margin-bottom: 30px;
}
/* Book Styling */
.book {
width: 332px;
height: 486px;
position: relative;
perspective: 1200px;
margin: 0 auto;
margin-left: 50%;
}
.page {
position: absolute;
width: 100%;
height: 100%;
top: 0;
left: 0;
transform-style: preserve-3d;
transform-origin: left center;
transition: all 0.7s cubic-bezier(0.645, 0.045, 0.355, 1);
border-radius: 5px 0 0 5px;
}
.front, .back {
position: absolute;
width: 100%;
height: 100%;
top: 0;
left: 0;
backface-visibility: hidden;
border-radius: 5px 0 0 5px;
display: flex;
flex-direction: column;
justify-content: center;
align-items: center;
padding: 15px;
font-size: 14px;
box-sizing: border-box;
box-shadow: inset 0 0 10px rgba(0,0,0,0.1);
}
.front {
background: linear-gradient(to right, #fefefe 95%, #f0f0f0 100%);
}
.back {
background: #f9f9f9;
transform: rotateY(180deg);
}
/* Page Content Styling */
.page-content {
width: 100%;
height: 100%;
overflow: hidden;
text-align: left;
}
.page-number {
position: absolute;
bottom: 10px;
font-size: 12px;
color: #777;
}
.front .page-number {
right: 15px;
}
.back .page-number {
left: 15px;
}
/* Cover Styling */
.cover {
background-image: url('https://cdn-uploads.huggingface.co/production/uploads/6671dd5203d6e8087aaf7ce5/F34iK8guTPplwEBCdwDAk.png');
background-size: cover;
background-repeat: no-repeat;
}
/* Hide radio buttons */
input[type="radio"] {
display: none;
}
.page.cover {
transform: rotateY(-180deg);
z-index: 0;
}
.page.page2 {
transform: rotateY(0deg);
z-index: 3;
}
.page.page3 {
transform: rotateY(0deg);
z-index: 2;
}
/* When cover radio is selected, show front cover */
#cover-radio:checked ~ .book .page.cover {
transform: rotateY(0deg);
z-index: 4;
}
/* When page1 radio is selected, turn cover to table of contents and show page1 */
#page1-radio:checked ~ .book .page.cover {
transform: rotateY(-180deg);
z-index: 0;
}
/* When page2 radio is selected, turn page1 and show page2 */
#page2-radio:checked ~ .book .page.cover {
transform: rotateY(-180deg);
z-index: 0;
}
#page2-radio:checked ~ .book .page.page2 {
transform: rotateY(-180deg);
z-index: 1;
}
/* When page3 radio is selected, turn page2 and show page3 */
#page3-radio:checked ~ .book .page.cover {
transform: rotateY(-180deg);
z-index: 0;
}
#page3-radio:checked ~ .book .page.page2 {
transform: rotateY(-180deg);
z-index: 1;
}
#page3-radio:checked ~ .book .page.page3 {
transform: rotateY(-180deg);
z-index: 2;
}
#page2-radio:checked ~ .book .page.cover {
transition-delay: 0.1s;
}
#page2-radio:checked ~ .book .page.page2 {
transition-delay: 0.2s;
}
#page2-radio:checked ~ .book .page.page3 {
transition-delay: 0.2s;
}
#page3-radio:checked ~ .book .page.page2 {
transition-delay: 0.1s;
}
#page3-radio:checked ~ .book .page.page3 {
transition-delay: 0.2s; /* Slightly longer delay for the 3rd page */
}
#cover-radio:checked ~ .book .page.cover {
transition-delay: 0.3s; /* Slightly longer delay for the 3rd page */
}
#cover-radio:checked ~ .book .page.page2 {
transition-delay: 0.2s;
}
#cover-radio:checked ~ .book .page.page3 {
transition-delay: 0.1s;
}
#page1-radio:checked ~ .book .page.page2 {
transition-delay: 0.2s;
}
#page1-radio:checked ~ .book .page.page3 {
transition-delay: 0.1s;
}
/* Control Dots */
.slider-controls {
margin-top: 30px;
display: flex;
justify-content: center;
gap: 15px;
}
.slider-controls label {
display: inline-block;
width: 12px;
height: 12px;
border-radius: 50%;
background: #bdc3c7;
cursor: pointer;
transition: background 0.3s;
}
.slider-controls label:hover {
background: #7f8c8d;
}
/* Back cover styling */
.back-cover {
background: #2c3e50;
}
.cover-back {
background: #2c3e50;
}
blockquote {
font-size: 14px;
padding: 0px;
padding-left: 7px;
margin: 0px;
margin-bottom: 4px;
}
.page-theme-black {
color: white;
background: #000000;
background: linear-gradient(0deg,rgba(0, 0, 0, 1) 0%, rgba(33, 37, 40, 1) 75%, rgba(89, 73, 82, 1) 86%, rgba(33, 37, 40, 1) 95%);
}
.page-theme-black h2 {
color: white;
}
.page-theme-black i {
color: #F7D9BC;
}
.page-theme-violet {
color: white;
background: #6d2dad;
background: radial-gradient(circle,rgba(109, 45, 173, 1) 0%, rgba(0, 0, 0, 1) 100%);
}
.page-theme-violet h2 {
color: #9d3ffb;
}
.page-theme-violet div {
color: #C9CFD5;
}
.page-theme-violet b {
color: white;
}
.page-theme-subtle {
background: #e8acac;
background: linear-gradient(90deg,rgba(232, 172, 172, 1) 0%, rgba(154, 193, 230, 1) 100%);
color: black;
}
.page-theme-subtle h2 {
color: black;
}
.page-theme-subtle li::marker {
color: #C72020;
}
.page-theme-subtle span {
color: #2A2F34;
}
.page-theme-subtle b {
color: black;
}
.page-theme-subtle .page-number {
color: black;
}
.page-theme-blue {
background: #090979;
background: radial-gradient(circle,rgba(9, 9, 121, 1) 70%, rgba(154, 193, 230, 1) 100%);
color: white;
}
.page-theme-blue h2 {
color: yellow;
}
.page-theme-blue blockquote {
color: white;
background-color: #70809050;
}
.page-theme-blue footer {
color: yellow;
}
.page-theme-blue .page-number {
color: white;
}
.block-theme-green {
background-color: #8BB06F;
border-radius: 10px;
color: black;
}
.block-theme-green h2 {
padding-left: 10px;
padding-bottom: 0px;
margin-bottom: 0px;
color: black;
}
.block-theme-green table {
margin-bottom: 5px;
}
.block-theme-plum {
background-color: #72366A;
border-radius: 10px;
color: white;
padding-left: 10px;
}
.block-theme-plum h2 {
margin-top: 5px;
padding-bottom: 0px;
margin-bottom: 0px;
color: white;
}
</style>
<div class="container">
<input type="radio" name="slider" id="cover-radio" checked>
<input type="radio" name="slider" id="page1-radio">
<input type="radio" name="slider" id="page2-radio">
<input type="radio" name="slider" id="page3-radio">
<div class="book">
<div class="page front">
<div class="page-content">
<h2>ACKNOWLEDGEMENTS</h2>
<ul>
<li>Mad lads, who provided feedback</li>
<li><a href="https://huggingface.co/yamatazen" target="_blank">yamatazen</a> - For high quality model merges</li>
<li>OG model authors - For making cool models</li>
<li><a href="https://huggingface.co/arcee-ai" target="_blank">Arcee AI</a> - For making <a href="https://github.com/arcee-ai/mergekit" target="_blank">mergekit</a></li>
<li><a href="https://huggingface.co/mradermacher" target="_blank">Team mradermacher</a> - For awesome quants</li>
<li><a href="https://huggingface.co/DeathGodlike" target="_blank">DeathGodlike</a> - For awesome quants in EXL3</li>
<li>Model mergers: <a href="https://huggingface.co/MrRikyz" target="_blank">Rickyz</a>, <a href="https://huggingface.co/Vortex5" target="_blank">Vortex</a>, <a href="https://huggingface.co/DreadPoor" target="_blank">Edward Eagley</a></li>
<li>You, for trying out my models</li>
</ul>
<div class="page-number">6</div>
</div>
</div>
<div class="page page3">
<div class="front page-theme-blue">
<div class="page-content">
<h2>WHAT USERS ARE SAYING</h2>
<blockquote cite="https://www.reddit.com/r/SillyTavernAI/comments/1nn5od9/comment/ngoo92d/">
Ive been using KSR for a week now, and I really like it. It is very creative and has brought together story threads in a natural way, even older ones at large context. I really like it so far.
<footer><cite>Pacoeltaco</cite></footer>
</blockquote>
<blockquote cite="https://www.reddit.com/r/SillyTavernAI/comments/1nb6wze/comment/ne88b3w/">
Ive had good results so far with my testing, smart(for 12b), follows prompts and uses things from the character card. the only thing I found negative so far is that it likes em dashes
<footer><cite>Background-Ad-5398</cite></footer>
</blockquote>
<blockquote cite="https://huggingface.co/Retreatcost/KansenSakura-Eclipse-RP-12b/discussions/1?not-for-all-audiences=true#68cd0eefd72cf1bf50e89b75">
I've been comparing this against a couple finetunes and this model's balance is impressive. I hope to see more soon!
<footer><cite>gggrandma1990</cite></footer>
</blockquote>
</div>
<div class="page-number">4</div>
</div>
<div class="back">
<div class="page-content">
<div class="block-theme-green">
<h2>SYSTEM REQUIREMENTS</h2>
<table>
<tr>
<td><b>Minimum<b></td>
<td><b>Recommended<b></td>
</tr>
<tr>
<td><b>Inference:</b> CPU<br><b>Memory:</b> 12GB RAM<br><b>Context Window:</b> 8K Tokens<br><b>Temp:</b> 0.65</td>
<td><b>Inference:</b> GPU<br><b>Memory:</b> 16GB+ VRAM<br><b>Context Window:</b> 16K Tokens<br><b>Temp:</b> 0.80<br><b>Courage:</b> High</td>
</tr>
</table>
</div>
<div class="block-theme-plum">
<h2>RECOMMENDED SETTINGS</h2>
<span>
<b>Temp</b>: 0.65-0.8 | <b>RepPen</b>: 1.05<br>
<b>TOP_P</b>: 0.95 | <b>MIN_P</b>: 0.05 | <b>TOP_K</b>: 0<br>
<b>Template Format</b>: ChatML | <b>Ctxt</b>: 16K
</span>
</div>
</div>
<div class="page-number">5</div>
</div>
</div>
<div class="page page2">
<div class="front page-theme-violet">
<div class="page-content">
<h2>CORE FEATURES</h2>
<ul>
<li><b>ADVANCED PSYCHOLOGICAL PROFILING</b><div>Characters feel realer than real - with all the darkness that implies</div></li>
<li><b>DREAD ATMOSPHERIC SYSTEMS</b><div>Environments that breathe, bleed, and remember your fears</div></li>
<li><b>EMOTIONAL EROSION ENGINE</b><div>Watch characters unravel under psychological pressure</div></li>
<li><b>UNFILTERED NARRATIVE DEPTH</b><div>Darker than Eclipse, more psychologically intense than Radiance</div></li>
</ul>
</div>
<div class="page-number">2</div>
</div>
<div class="back page-theme-subtle">
<div class="page-content">
<h2>FAQ</h2>
<ul>
<li><b>Q: IS THIS MODEL BETTER THAN X?</b><br><span>A: <b>Different weapon, different war.</b> Erosion specializes in psychological depth and darker narratives. It's smarter than previous versions, but choose your tool for the mission.</span></li>
</ul>
<ul>
<li><b>Q: WHEN'S THE NEXT VERSION COMING?</b><br><span>A: Currently experimenting with <b>finetuning</b> tech. Expect experimental merges while we develop the next major build.<span></li>
</ul>
<ul>
<li><b>Q: PLANNING OTHER ARCHITECTURES?</b><br><span>A: <b>Affirmative.</b> Reconnaissance underway for larger models - Mistral Small and Qwen3 are potential candidates for future deployments.</span></li>
</ul>
</div>
<div class="page-number">3</div>
</div>
</div>
<div class="page cover">
<div class="back page-theme-black">
<div class="page-content">
<h2>THE ULTIMATE RP ENGINE JUST GOT DARKER</h2>
<p>
<i>They told us to extract the sakura essence. They never told us it would remember.</i>
</p>
<p>
<b>KansenSakura: Erosion</b> represents the final stage of neural corruption - where beauty becomes a weapon and every memory turns to poison.
</p>
<p>
<i>Some blossoms grow best in poisoned soil. Welcome to the garden.</i>
</p>
<center>
<div>
<div>Psychological Horror | Intense Themes</div>
<div>Complex Narratives | NSFW</div>
<div>KansenSakura Project | Containment Failed</div>
<div><b>RATED M for MATURE</b></div>
</div>
</center>
</div>
<div class="page-number">1</div>
</div>
</div>
</div>
</div>
<center>
<div class="slider-controls">
<label for="cover-radio"></label>
<label for="page1-radio"></label>
<label for="page2-radio"></label>
<label for="page3-radio"></label>
</div>
<div style="margin-top: 20px; color: #666; font-size: 14px;">
<p>Click the dots to navigate through the book.</p>
</div>
<audio controls src="https://cdn-uploads.huggingface.co/production/uploads/6671dd5203d6e8087aaf7ce5/4Kt08OL1g6OrbCu6SAFQe.mpga"></audio>
</center>
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
<details>
<summary>Merge Details</summary>
### Merge Method
This model was merged using the [Multi-SLERP](https://goddard.blog/posts/multislerp-wow-what-a-cool-idea) merge method.
### Models Merged
The following models were included in the merge:
* [Retreatcost/Irix-mpf-stock](https://huggingface.co/Retreatcost/Irix-mpf-stock)
* [Retreatcost/Forgotten-directive-Neon-stock](https://huggingface.co/Retreatcost/Forgotten-directive-Neon-stock)
* [Retreatcost/Lorablated-w2bb-psy-della](https://huggingface.co/Retreatcost/Lorablated-w2bb-psy-della)
* [mistralai/Mistral-Nemo-Base-2407](https://huggingface.co/mistralai/Mistral-Nemo-Base-2407)
* [yamatazen/EtherealAurora-12B-Lorablated](https://huggingface.co/yamatazen/EtherealAurora-12B-Lorablated)
* [Retreatcost/Shisa-K-sakurization](https://huggingface.co/Retreatcost/Shisa-K-sakurization)
* [Sicarius-Prototyping/Impish_Longtail_12B](https://huggingface.co/Sicarius-Prototyping/Impish_Longtail_12B)
* [SuperbEmphasis/MN-12b-RP-Ink-RP-Longform](https://huggingface.co/SuperbEmphasis/MN-12b-RP-Ink-RP-Longform)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
merge_method: multislerp
models:
- model: yamatazen/EtherealAurora-12B-Lorablated
parameters:
weight: [1.000, 1.000, 1.000, 1.000, 0.968, 0.744, 0.256, 0.030, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.0225, 0.1275, 0.15]
- model: Retreatcost/Shisa-K-sakurization
parameters:
weight: [0.000, 0.000, 0.000, 0.030, 0.256, 0.744, 0.968, 1.000, 1.000, 1.000, 1.000, 0.968, 0.744, 0.256, 0.030, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.015, 0.085, 0.10]
- model: ./impish-ink-long
parameters:
weight: [0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.030, 0.256, 0.744, 0.968, 1.000, 1.000, 1.000, 0.968, 0.744, 0.256, 0.030, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.0075, 0.0425, 0.05]
- model: Retreatcost/Lorablated-w2bb-psy-della
parameters:
weight: [0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.030, 0.256, 0.744, 0.968, 1.000, 1.000, 1.000, 0.968, 0.744, 0.256, 0.030, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000]
- model: Retreatcost/Forgotten-directive-Neon-stock
parameters:
weight: [0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.030, 0.256, 0.744, 0.968, 1.000, 1.000, 1.000, 0.968, 0.744, 0.256, 0.030, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000]
- model: Retreatcost/Irix-mpf-stock
parameters:
weight: [0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.030, 0.256, 0.744, 0.968, 1.000, 1.000, 1.000, 1.000, 1.000, 0.850, 0.150, 0.000]
- model: ./retokenized_NB
parameters:
weight: [0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.105, 0.595, 0.700]
dtype: bfloat16
parameters:
normalize: true
tokenizer_source: Retreatcost/KansenSakura-Radiance-RP-12b
```
</details>
[ [**BACK**](#) ] [ [**MAIN MENU**](https://huggingface.co/Retreatcost) ] [ [ **QUIT** ](/)]