WizardLM-Uncensored-SuperCOT-StoryTelling-30B-GPTQ
| Property | Value |
|---|---|
| Parameter Count | 30B |
| Model Type | GPTQ-Quantized LLaMA |
| License | Other |
| Author | TheBloke (Quantized) / YellowRoseCx (Original) |
What is WizardLM-Uncensored-SuperCOT-StoryTelling-30B-GPTQ?
This is a sophisticated quantized language model that combines three key capabilities: uncensored text generation, chain-of-thought reasoning, and advanced storytelling. It's based on the WizardLM architecture and has been optimized through GPTQ quantization to run efficiently while maintaining high performance.
Implementation Details
The model is available in multiple GPTQ configurations, ranging from 3-bit to 8-bit quantization with various group sizes. The main branch offers a 4-bit implementation with Act Order enabled and no group size, optimizing for lower VRAM requirements while maintaining quality.
- Multiple quantization options (3-bit, 4-bit, 8-bit)
- Supports different group sizes (32g, 64g, 128g)
- Compatible with ExLlama for 4-bit versions
- Optimized with wikitext dataset for quantization
Core Capabilities
- Advanced storytelling and narrative generation
- Enhanced reasoning through Chain-of-Thought methodology
- Uncensored text generation capabilities
- Flexible deployment options for different hardware configurations
Frequently Asked Questions
Q: What makes this model unique?
This model stands out due to its triple-merge architecture combining uncensored capabilities, chain-of-thought reasoning, and storytelling abilities, making it particularly versatile for creative and analytical tasks.
Q: What are the recommended use cases?
The model excels at creative writing, storytelling, complex reasoning tasks, and scenarios requiring detailed explanations or narrative generation. It's particularly suitable for applications needing both creative and analytical capabilities.
