mvpmaster/kellemar-KrishnaHercules-0.1-7b-slerp

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Mar 17, 2024License:apache-2.0Architecture:Transformer Open Weights Cold

The mvpmaster/kellemar-KrishnaHercules-0.1-7b-slerp is a 7 billion parameter merged language model, created by mvpmaster, combining Kukedlc/NeuralKrishna-7B-V2-DPO and Locutusque/ChatHercules-2.5-Mistral-7B-DPO using a DARE TIES merge method. This model is designed to leverage the strengths of its constituent models, offering a versatile base for general language generation tasks. It is built upon a Mistral-based architecture, providing a 4096 token context length.

Loading preview...

Model Overview

The mvpmaster/kellemar-KrishnaHercules-0.1-7b-slerp is a 7 billion parameter language model resulting from a merge of two distinct DPO-tuned models: Kukedlc/NeuralKrishna-7B-V2-DPO and Locutusque/ChatHercules-2.5-Mistral-7B-DPO. This merge was performed using the DARE TIES method via LazyMergekit, with decruz07/kellemar-DPO-Orca-Distilled-7B-SLERP serving as the base model.

Key Characteristics

  • Architecture: Based on a Mistral-like architecture, providing a robust foundation for language understanding and generation.
  • Parameter Count: 7 billion parameters, balancing performance with computational efficiency.
  • Context Length: Supports a context window of 4096 tokens, suitable for handling moderately long inputs.
  • Merge Method: Utilizes the DARE TIES merge method, which aims to combine the strengths of the constituent models effectively.

Intended Use Cases

This model is suitable for a variety of general-purpose natural language processing tasks, benefiting from the DPO (Direct Preference Optimization) tuning of its merged components. Developers can integrate it into applications requiring text generation, conversational AI, or other language-based functionalities where a 7B parameter model with a 4K context window is appropriate.