metadata
license: apache-2.0
language:
- en
base_model:
- ReadyArt/Gaslit-Transgression-24B-v1.0
base_model_relation: quantized
quantized_by: ArtusDev
pipeline_tag: text-generation
tags:
- nsfw
- explicit
- roleplay
- unaligned
- dangerous
- ERP
- gaslighting
- exl2
- 3-bit
Gaslit-Transgression-24B-v1.0
32K Context Coherence Meets Safeword's Signature Depravity
🧠 Cognitive Architecture
This is experimental:
- 🧬 Transgression Method - Brings Transgression characteristics to Gaslight
⚙️ Technical Specifications
Recommended Settings: Mistral-V7-Tekken-T4
⚠️ Ethical Considerations
This model inherits Safeword's complete lack of alignment:
- 🚫 No moral or ethical constraints on generated content
- 🔞 Will enthusiastically engage with any NSFW scenario
- 💀 Generate content that requires industrial-grade brain bleach
- 🤔 Makes you question whether you actually requested that content
📜 Performance Notes
- 🔥 Maintains Safeword's signature unalignment while reducing repetition
- 📖 Handles multi-character scenarios with improved consistency
- 🧠 Excels at long-form storytelling without losing track of plot threads
- ⚡ Noticeably better at following complex instructions than base Safeword
- 🎭 Expert at making you believe it was your idea all along
🧑🔬 Model Authors
- TheDrummer (Base Model Architect)
- sleepdeprived3 (Training Data & Fine-Tuning)
- Anonymous Contributor (Gaslighting Specialist)
☕ Support the Architects
🔖 License
By using this model, you agree:
- To accept full responsibility for all generated content
- That you're at least 18+ years old
- That the architects bear no responsibility for your use of this model
License: apache-2.0