zarakiquemparte/hermes-kimiko-7b
The zarakiquemparte/hermes-kimiko-7b is a 7 billion parameter language model created by zarakiquemparte, built upon the Nous Hermes Llama2 7b base model and merged with the Kimiko LLama2 7B Lora. This model combines the strengths of its base components, offering a versatile foundation for various natural language processing tasks. It supports both Alpaca 2 and LimaRP instruction formats, making it adaptable for different conversational and prompt-based applications. With a context length of 4096 tokens, it is suitable for tasks requiring moderate input and output lengths.
Loading preview...
Model Overview
The zarakiquemparte/hermes-kimiko-7b is a 7 billion parameter language model developed by zarakiquemparte. It is a merged model, combining the Nous Hermes Llama2 7b as its base with the Kimiko LLama2 7B Lora (at 1.0 weight). This merging process was facilitated by a custom script, allowing for the integration of capabilities from both foundational models.
Key Features & Usage
This model is designed to be flexible in its instruction formats, supporting two common structures:
- Alpaca 2 format: Ideal for direct instruction-response interactions.
### Instruction: <prompt> ### Response: <leave a newline blank for model to respond> - LimaRP format: Suitable for role-playing and character-driven interactions, allowing for system and character card definitions.
<<SYSTEM>> <character card and system prompt> <<USER>> <prompt> <<AIBOT>> <leave a newline blank for model to respond>
Limitations
It is important to note that this model is not intended for providing factual information or advice. Users should exercise caution and verify any information generated by the model, as its primary purpose is not factual accuracy.
Reproducibility
The merging process used to create this model is transparent and can be reproduced using the provided script. Further details on the base models can be found via their respective links: Nous Hermes Llama2 7b and Kimiko LLama2 7B Lora.