l3utterfly/mistral-7b-v0.1-layla-v4-chatml
l3utterfly/mistral-7b-v0.1-layla-v4-chatml is a 7 billion parameter Mistral-based language model developed by l3utterfly, fine-tuned on the OpenHermes 2.5 dataset. This model is specifically optimized for multi-turn conversational interactions and character impersonation, making it suitable for applications requiring nuanced persona adoption. It features a 4096-token context length and is designed for English language processing.
Loading preview...
Overview
l3utterfly/mistral-7b-v0.1-layla-v4-chatml is a 7 billion parameter model built upon the Mistral 7B architecture, developed by l3utterfly and funded by Layla Network. It has been meticulously fine-tuned using the OpenHermes 2.5 dataset, with additional NSFW conversations from the Teatime dataset. The training data underwent specific preprocessing, including the removal of refusals and AI assistant mentions, and the splitting of multi-turn dialogues into distinct conversation records.
Key Capabilities
- Multi-turn Conversation: Optimized for engaging in extended, natural dialogues.
- Character Impersonation: Excels at adopting and maintaining specific character personalities and traits, as demonstrated by its prompt examples.
- English Language Processing: Primarily focused on English language tasks.
Good For
- Personal Assistants: Serves as the base model for Layla, an offline personal assistant, indicating its suitability for interactive agent roles.
- Role-playing Scenarios: Ideal for applications requiring the model to embody a specific persona or character.
- Interactive Applications: Well-suited for chatbots and conversational AI where maintaining context and character is crucial.