bunnycore/Starling-dolphin-E26-7B
Starling-dolphin-E26-7B is an experimental 7 billion parameter language model developed by bunnycore, created by merging several existing models. This model aims to handle a wide range of tasks, including instruction following and creative writing, leveraging its merged architecture for broad applicability. It is still under development, focusing on exploring the capabilities of merged models.
Loading preview...
Overview
Starling-dolphin-E26-7B is an experimental 7 billion parameter language model developed by bunnycore. It was created by merging multiple powerful models using the model_stock merging method, specifically leveraging mergekit. The base model for this merge is cognitivecomputations/dolphin-2.2.1-mistral-7b.
Key Capabilities
- Instruction Following: Designed to accurately interpret and execute user instructions.
- Creative Writing: Aims to generate diverse and imaginative text.
- Broad Task Handling: The merging approach is intended to equip the model with a wide range of general-purpose language understanding and generation abilities.
Good For
- Experimental Applications: Ideal for developers and researchers interested in exploring the performance and emergent properties of merged language models.
- General-Purpose Text Generation: Suitable for tasks requiring both structured instruction adherence and creative output.
- Development and Prototyping: As an experimental model, it serves as a strong foundation for further fine-tuning and application development.
Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.