darkc0de/XortronGlitched
darkc0de/XortronGlitched is a 24 billion parameter language model created by darkc0de, built upon the TroyDoesAI/BlackSheep-24B base model using the DELLA merge method. This model integrates components from several Mistral-based models, including Dolphin3.0-Mistral-24B and Cydonia-24B-v2. It is designed to combine the strengths of its constituent models, offering a versatile foundation for various generative AI tasks with a 32768 token context length.
Loading preview...
Overview
darkc0de/XortronGlitched is a 24 billion parameter language model developed by darkc0de, created through a sophisticated merge of several pre-trained models. It utilizes the DELLA merge method with TroyDoesAI/BlackSheep-24B serving as its foundational base model.
Key Components
This model is a composite of the following 24B parameter models, each contributing to its overall capabilities:
- cognitivecomputations/Dolphin3.0-Mistral-24B
- TheDrummer/Cydonia-24B-v2
- huihui-ai/Mistral-Small-24B-Instruct-2501-abliterated
- huihui-ai/Arcee-Blitz-abliterated
Merge Configuration
The merge process employed a specific configuration, assigning equal density and weight parameters (0.5) to each contributing model. The int8_mask parameter was enabled, and the model was configured with float16 dtype, indicating an optimization for efficiency while maintaining performance. This approach aims to synthesize the diverse strengths of its constituent models into a single, robust offering.
Potential Use Cases
Given its merged architecture, XortronGlitched is likely suitable for a broad range of applications that benefit from the combined knowledge and capabilities of its base models. Developers seeking a versatile 24B parameter model with a 32768 token context length, built from established Mistral-based foundations, may find this model particularly useful.