elinas/chronos-70b-v2
The elinas/chronos-70b-v2 is a 69 billion parameter Llama v2 Base model, developed by Elinas, primarily focused on chat, roleplay, and storywriting. It features significantly improved reasoning and logic capabilities, supporting a context length of up to 4096 tokens. This model excels at generating long, coherent text outputs and is designed for non-commercial use.
Loading preview...
Overview
elinas/chronos-70b-v2 is a 69 billion parameter model built upon the Llama v2 Base architecture, developed by Elinas. It is specifically optimized for chat, roleplay, and storywriting applications, demonstrating enhanced reasoning and logical coherence. The model supports a context length of up to 4096 tokens and is capable of generating very long, coherent text outputs.
Key Capabilities
- Chat and Roleplay: Designed for interactive conversational scenarios.
- Storywriting: Excels at generating extended, coherent narratives.
- Improved Reasoning: Features significantly enhanced logical processing compared to its base model.
- Long Output Generation: Capable of producing lengthy and coherent text.
- Alpaca Formatting: Requires Alpaca instruction formatting for optimal performance.
Usage and Licensing
This model is released under a strict non-commercial (cc-by-nc-4.0) license, which takes precedence over the LLAMA 2 COMMUNITY LICENSE AGREEMENT. Commercial use requires direct contact with Elinas. Users are advised to utilize Alpaca formatting (### Instruction:\nYour instruction or question here.\n### Response:) for best results. Quantized versions are available for consumer GPU usage, provided by TheBloke.