sequelbox/Llama2-70B-SpellBlade
TEXT GENERATIONConcurrency Cost:4Model Size:69BQuant:FP8Ctx Length:32kPublished:Dec 24, 2023License:apache-2.0Architecture:Transformer0.0K Open Weights Cold
sequelbox/Llama2-70B-SpellBlade is a 69 billion parameter Llama 2-based model with a 32768 token context length, fine-tuned for enhanced conversational quality and technical capabilities. It aims to improve general chat performance and technical proficiency over the base Llama 2 architecture. This model is primarily offered for reference, with most training data utilizing the [INST][/INST] chat format.
Loading preview...
Model Overview
sequelbox/Llama2-70B-SpellBlade is a 69 billion parameter model built upon the Llama 2 architecture, featuring a 32768 token context length. It has been fine-tuned to enhance both conversational quality and technical capabilities, aiming to provide an upgrade over the base Llama 2 model in these areas.
Key Characteristics
- Architecture: Based on Llama 2, a widely recognized open-source large language model.
- Parameter Count: 69 billion parameters, indicating a substantial model size for complex tasks.
- Context Length: Supports a 32768 token context window, allowing for processing longer inputs and maintaining conversational coherence over extended interactions.
- Training Data: Primarily trained using the
[INST][/INST]chat format, optimizing it for instruction-following and conversational exchanges. - Purpose: Designed to improve general chat performance and supplement technical proficiency.
Usage Considerations
- Performance: The model performs solidly as-is, but user satisfaction is expected to be optimized with further fine-tuning for specific applications.
- Legacy Status: This model is considered a 'legacy model' and is offered primarily for reference. For general use, the developer recommends Llama 3 as a more current alternative.