kodonho/SolarM-SakuraSolar-SLERP

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:10.7BQuant:FP8Ctx Length:4kPublished:Jan 12, 2024License:cc-by-nc-4.0Architecture:Transformer0.0K Open Weights Warm

kodonho/SolarM-SakuraSolar-SLERP is a 10.7 billion parameter English mixed language model created by kodonho. This model is a gradient SLERP merge of DopeorNope/SOLARC-M-10.7B and kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v2, designed to combine their respective strengths. It is suitable for general text generation tasks, leveraging its merged architecture for improved performance.

Loading preview...

Overview

kodonho/SolarM-SakuraSolar-SLERP is a 10.7 billion parameter English language model developed by kodonho. This model is a result of a gradient SLERP (Spherical Linear Interpolation) merge, combining the capabilities of two distinct base models: DopeorNope/SOLARC-M-10.7B and kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v2.

Key Characteristics

  • Merged Architecture: Utilizes a gradient SLERP technique to blend the strengths of two pre-existing models, aiming for enhanced performance across various tasks.
  • English Language Focus: Primarily designed for English language understanding and generation.
  • Parameter Count: Operates with 10.7 billion parameters, offering a balance between computational efficiency and model capability.

Use Cases

This model is suitable for a range of general-purpose natural language processing applications where a robust English language model is required. Its merged heritage suggests potential for improved reasoning or instruction-following capabilities, depending on the specific strengths inherited from its base models.