Model Overview
DS-Archive/Nous-Hermes-Llama2-13b-Kimiko-Lora-Merged is a 13 billion parameter model built upon the Llama 2 architecture. It is a merge of two distinct components: the Nous Hermes Llama 2 13b base model and the Kimiko Lora adapter. The primary objective of this merge was to create a model that excels in both general instruction following and detailed character roleplay.
Key Capabilities
- Combined Strengths: Integrates the robust reasoning and instruction-following abilities of Nous Hermes Llama 2 with the specialized character roleplay capabilities derived from the Kimiko Lora.
- Flexible Prompting: Supports two distinct instruction formats: the Alpaca format from the Nous Hermes Llama 2 base model and the specific
<<SYSTEM>>, <<USER>>, <<AIBOT>> format from the Kimiko Lora. - Context Length: Operates with a context window of 4096 tokens.
Intended Use Cases
This model is particularly well-suited for applications requiring a blend of general conversational intelligence and the ability to maintain consistent character personas. It can be used for:
- Character-driven Roleplay: Generating responses that adhere to specific character traits and backstories.
- Instruction-based Interactions: Responding to direct instructions while maintaining a character context.
Limitations and Biases
Users should be aware that the model may exhibit biases present in niche roleplaying communities, in addition to those inherited from its base model. It is not designed or intended for providing factual information or professional advice.