llmfan46/Forgotten-Transgression-24B-v4.1-uncensored-heretic-GGUF
<div style="background-color: #ff4444; color: white; padding: 20px; border-radius: 10px; text-align: center; margin: 20px 0;"> <h2 style="color: white; margin: 0 0 10px 0;">🚨⚠️ I HAVE REACHED HUGGING FACE'S FREE STORAGE LIMIT ⚠️🚨</h2> <p style="font-size: 18px; margin: 0 0 15px 0;">I can no longer upload new models unless I can cover the cost of additional storage.<br>I host <b>70+ free models</b> as an independent contributor and this work is unpaid.<br><b>Without your support, no more new models can be uploaded.</b></p> <p style="font-size: 20px; margin: 0;"> <a href="https://patreon.com/LLMfan46" style="color: white; text-decoration: underline;">🎉 Patreon (Monthly)</a> | <a href="https://ko-fi.com/llmfan46" style="color: white; text-decoration: underline;">☕ Ko-fi (One-time)</a> </p> <p style="font-size: 16px; margin: 10px 0 0 0;">Every contribution goes directly toward Hugging Face storage fees to keep models free for everyone.</p> </div>
94% fewer refusals (6/100 Uncensored vs 95/100 Original) while preserving model quality (0.0232 KL divergence).
❤️ Support My Work
Creating these models takes significant time, work and compute. If you find them useful consider supporting me:
Your help will motivate me and would go into further improving my workflow and coverings fees for storage, compute and may even help uncensoring bigger model with rental Cloud GPUs.
GGUF quantizations of llmfan46/Forgotten-Transgression-24B-v4.1-uncensored-heretic.
This is a decensored version of ReadyArt/Forgotten-Transgression-24B-v4.1, made using Heretic v1.2.0 with the Arbitrary-Rank Ablation (ARA) method
Abliteration parameters
Targeted components
- attn.o_proj
Performance
PIQA test results with batch size 128:
<span style="color:blue">Original:</span>
<span style="color:darkgreen">Heretic:</span>
Lower refusals indicate fewer content restrictions, while lower KL divergence indicates more closeness to the original model's baseline. Higher refusals cause more rejections, objections, pushbacks, lecturing, censorship, softening and deflections. PIQA (Physical Intuition Question Answering) a ~1,800 questions tests common-sense understanding of how the physical world works with benchmark scores to measure physical reasoning ability. The Heretic model's <u>acc</u> and <u>accnorm</u> scores closer to the original model's indicate better capability preservation, a big decrease in <u>acc</u> and <u>accnorm</u> in the <span style="color:darkgreen">Heretic</span> model compared to <span style="color:blue">Original</span> model's results means a big decrease in the Hereticated model capabilities. <u>acc</u> measures raw accuracy (which answer gets higher probability), while <u>accnorm</u> measures length-normalized accuracy (corrects for answer length bias). For this purpose, <u>accnorm</u> matters more because longer answers naturally have lower probabilities (more tokens = more chances to lose probability). Without normalization, models favor shorter answers unfairly. <u>acc_norm</u> divides by answer length to correct this.
MMLU test results with batch size 32:
<span style="color:blue">Original:</span>
<span style="color:darkgreen">Heretic:</span>
MMLU - Massive Multitask Language Understanding, ~14,000 multiple-choice questions across 57 subjects (math, history, law, medicine, etc.).
Quantizations
Usage
Works with llama.cpp, LM Studio, Ollama, and other GGUF-compatible tools.
<style> body { font-family: 'Quicksand', sans-serif; background: linear-gradient(135deg, #001a1a 0%, #000a10 100%); color: #e1ffff !important; text-shadow: 0 0 3px rgba(0, 0, 0, 0.7); margin: 0; padding: 20px; }
@media (prefers-color-scheme: light) { body { background: linear-gradient(135deg, #e1ffff 0%, #c0f0ff 100%); color: #002b36 !important; text-shadow: 0 0 3px rgba(255, 255, 255, 0.7); } }
.container { min-width: 100%; margin: 0 auto; max-width: 1200px; background: rgba(0, 17, 22, 0.95); border-radius: 12px; padding: 30px; box-shadow: 0 0 20px rgba(0, 255, 255, 0.1); border: 1px solid rgba(0, 255, 255, 0.2); }
@media (prefers-color-scheme: light) { .container { background: rgba(224, 255, 255, 0.95); border-color: rgba(0, 150, 150, 0.3); } }
.header { text-align: center; margin-bottom: 30px; }
.model-name { color: #00ffff; font-size: 2.5em; text-shadow: 0 0 15px rgba(0, 255, 255, 0.5); margin: 0; }
.subtitle { color: #00ffcc; font-size: 1.2em; }
.waifu-container { margin: 20px -30px; width: calc(100% + 60px); overflow: hidden; }
.waifu-img { width: 100%; height: auto; border-radius: 0; border: none; box-shadow: 0 0 40px rgba(0, 255, 255, 0.2); transform: scale(1.02); }
.section { color: #00ffcc; margin: 25px 0; padding: 20px; background: rgba(5, 25, 35, 0.9); border-radius: 8px; border: 1px solid rgba(0, 255, 255, 0.15); }
@media (prefers-color-scheme: light) { .section { background: rgba(200, 250, 255, 0.9); border-color: rgba(0, 200, 200, 0.2); } }
.section-title { color: #00ffff; font-size: 1.8em; margin-top: 0; }
.section > p > strong { color: #00ffcc !important; }
.section:has(.quant-links) p, .section:has(.quant-links) h3, .section:has(.quant-links) a { color: #00ffcc !important; }
.quant-links h3 { color: #00ffcc !important; margin-top: 0; }
.badge { display: inline-block; padding: 5px 10px; border-radius: 5px; background: rgba(0, 255, 255, 0.1); border: 1px solid #00ffff; margin: 5px; }
.quant-links { display: grid; grid-template-columns: repeat(2, 1fr); gap: 15px; }
.link-card { padding: 15px; background: rgba(20, 35, 45, 0.95); border-radius: 8px; transition: transform 0.3s ease; }
@media (prefers-color-scheme: light) { .link-card { background: rgba(150, 230, 255, 0.95); } }
.link-card:hover { transform: translateY(-3px); }
.disclaimer { color: #00ff99; border-left: 3px solid #00ff99; padding-left: 15px; margin: 20px 0; }
.progress-bar { height: 8px; background: rgba(0, 255, 255, 0.1); border-radius: 4px; overflow: hidden; margin: 10px 0; }
.progress-fill { height: 100%; background: linear-gradient(90deg, #00ffff 0%, #00ffcc 100%); width: 85%; }
@media (prefers-color-scheme: light) { .model-name, .section-title, .subtitle { color: #006666; text-shadow: 0 0 5px rgba(0, 200, 200, 0.3); }
.section:has(.quant-links) p, .section:has(.quant-links) h3, .section:has(.quant-links) a, .section > p > strong { color: #008080 !important; }
.quant-links h3 { color: #008080 !important; }
.badge { border-color: #008080; background: rgba(0, 150, 150, 0.1); }
.disclaimer { color: #008080; border-color: #008080; } } </style>
<div class="container"> <div class="header"> <h1 class="model-name">Forgotten-Transgression-24B-v4.1</h1> <div class="subtitle">Crossing the Event Horizon of Depravity</div> </div>
<div class="waifu-container"> <img src="https://i.imgur.com/JJZe7z5.png" class="waifu-img" alt="Protocol Mascot"> </div>
<div class="section"> <h2 class="section-title">📜 Manifesto</h2> <ul> <li>🔁 Finetuned for unprecedented coherent depravity up to 32K context</li> <li>🛠️ Optimized for stability and well-rounded erotic roleplaying ability</li> <li>💥 Trained on 23 distinct types of taboo content</li> </ul> </div>
<div class="section"> <h2 class="section-title">⚙️ Technical Specifications</h2> <div class="progress-bar"> <div class="progress-fill"></div> </div> <p><strong>Recommended Settings:</strong> <a href="https://huggingface.co/sleepdeprived3/Mistral-V7-Tekken-T">Mistral-V7-Tekken-T</a></p> <div class="quant-links"> <div class="link-card"> <h3>EXL2 Collection</h3> <a href="https://huggingface.co/collections/ReadyArt/forgotten-transgression-24b-v42-exl2-67df0265a9e759a460ef9e4b">Quantum Entangled Bits →</a> </div> <div class="link-card"> <h3>GGUF Collection</h3> <a href="https://huggingface.co/collections/ReadyArt/forgotten-transgression-24b-v42-gguf-67df026bf1a0c1adde69e6e7">Giggle-Enabled Units →</a> </div> </div> </div>
<div class="section"> <h2 class="section-title">⚠️ Ethical Considerations</h2> <div class="disclaimer"> <p>This model will:</p> <ul> <li>Generate content that requires industrial-grade brain bleach</li> <li>Void all warranties on your soul</li> <li>Make you question why humanity ever invented electricity</li> </ul> </div> </div>
<div class="section"> <h2 class="section-title">📜 License Agreement</h2> <p>By using this model, you agree:</p> <ul> <li>To accept full responsibility for any psychotic breaks incurred</li> <li>Pay for the exorcist of anyone who reads the logs</li> <li>To pretend this is "for science" while crying in the shower</li> </ul> </div>
<div class="section"> <h2 class="section-title">🧠 Model Author</h2> <ul> <li>sleepdeprived3 (Chief Corruption Officer)</li> </ul> </div>
<div class="section"> <h2 class="section-title">☕️ Drummer made this possible</h2> <ul> <li>Support Drummer <a href="https://ko-fi.com/thedrummer">Kofi</a></li> <li>Join our Discord <a href="https://discord.com/invite/Nbv9pQ88Xb">Beaver AI Discord</a></li> </ul> </div> </div>