SakanaAI/EvoLLM-JP-v1-7B
EvoLLM-JP-v1-7B is a 7 billion parameter experimental general-purpose Japanese autoregressive language model developed by Sakana AI. It was created using an Evolutionary Model Merge method, combining Shisa Gamma 7B v1, WizardMath 7B V1.1, and Abel 7B 002. This model is specifically designed for research and development in Japanese language processing, leveraging its unique merging approach to achieve its capabilities.
Loading preview...
Overview
SakanaAI/EvoLLM-JP-v1-7B is a 7 billion parameter experimental general-purpose Japanese language model developed by Sakana AI. This model is notable for its creation via an Evolutionary Model Merge method, combining the strengths of three distinct base models: Shisa Gamma 7B v1, WizardMath 7B V1.1, and Abel 7B 002. The methodology behind its development is detailed in their paper and blog post.
Key Characteristics
- Language Focus: Primarily designed for Japanese language processing.
- Development Method: Utilizes an innovative Evolutionary Model Merge technique.
- Base Models: Merges capabilities from Shisa Gamma 7B v1, WizardMath 7B V1.1, and Abel 7B 002.
- License: Distributed under the MICROSOFT RESEARCH LICENSE TERMS, due to the inclusion of WizardMath.
Intended Use
This model is provided for research and development purposes only and is considered an experimental prototype. It is not intended for commercial use or deployment in mission-critical environments. Users should be aware of its experimental nature and use it at their own discretion.