Eric111/MarcoHermes

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Feb 9, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

MarcoHermes is a 7 billion parameter language model created by Eric111, formed by merging AtAndDev/CapybaraMarcoroni-7B and eren23/DistilHermes-2.5-Mistral-7B using the mergekit tool. This model leverages the strengths of its constituent models, offering a 4096-token context length. It is designed to combine the capabilities of its base models, making it suitable for general-purpose language generation tasks.

Loading preview...

Model Overview

MarcoHermes is a 7 billion parameter language model developed by Eric111. It is a merged model, created using the mergekit tool, combining two distinct base models: AtAndDev/CapybaraMarcoroni-7B and eren23/DistilHermes-2.5-Mistral-7B.

Key Characteristics

  • Architecture: A blend of two 7B parameter models, leveraging their respective strengths.
  • Merge Method: Utilizes the slerp (spherical linear interpolation) merge method, with specific parameter weighting applied to self-attention and MLP layers to optimize performance.
  • Context Length: Supports a context window of 4096 tokens.
  • Data Type: Configured to use bfloat16 for efficient computation.

Use Cases

This model is suitable for general language generation tasks, benefiting from the combined capabilities of its merged components. Developers looking for a 7B model that integrates features from both CapybaraMarcoroni and DistilHermes-2.5-Mistral may find MarcoHermes a compelling option for various NLP applications.