Chronos-Gold-12B-1.0
| Property | Value |
|---|---|
| Parameter Count | 12.2B |
| Base Model | Mistral-Nemo-Base-2407 |
| License | Apache 2.0 |
| Context Length | 16384 tokens |
| Tensor Type | BF16 |
What is Chronos-Gold-12B-1.0?
Chronos-Gold-12B-1.0 is an advanced language model built on the Mistral-Nemo architecture, specifically designed for creative writing, roleplay, and general conversational tasks. This model represents a significant evolution in AI text generation, capable of producing up to 2250 tokens in a single sequence while maintaining coherence and context awareness.
Implementation Details
The model utilizes the ChatML format for interactions and has been extensively modified from its Mistral-Nemo base to achieve enhanced prompt adherence and coherence. It features optimized sampling settings with recommended temperature of 0.7 and presence penalty of 1.0 for optimal performance.
- Trained at 16384 sequence length
- Supports various character card formats
- Multiple quantization options available (GGUF, Exllama2, FP8)
- Incorporates advanced prompt following capabilities
Core Capabilities
- General chatbot functionality
- Creative writing and storytelling
- Character-based roleplay
- Long-form content generation
- Consistent performance in standardized evaluations (31.66% on IFEval, 35.91% on BBH)
Frequently Asked Questions
Q: What makes this model unique?
The model's unique strength lies in its ability to maintain coherence across long sequences while providing enhanced prompt adherence, making it particularly suitable for creative writing and roleplay scenarios. It also supports multiple character card formats and offers various quantization options for different deployment needs.
Q: What are the recommended use cases?
The model excels in creative writing, storytelling, roleplay scenarios, and general conversational tasks. It's particularly well-suited for applications requiring extended context understanding and consistent character portrayal.
