arcee-ai/Saul-Legal-Calme-Instruct
Saul-Legal-Calme-Instruct is a 7 billion parameter instruction-tuned language model created by arcee-ai, formed by merging MaziyarPanahi/Calme-7B-Instruct-v0.1.1 and Equall/Saul-Instruct-v1. This model is designed for general instruction following, leveraging the combined strengths of its constituent models. It offers a 4096-token context length, making it suitable for a variety of natural language processing tasks.
Loading preview...
Saul-Legal-Calme-Instruct Overview
Saul-Legal-Calme-Instruct is a 7 billion parameter instruction-tuned language model developed by arcee-ai. It was created using mergekit by combining two distinct models: MaziyarPanahi/Calme-7B-Instruct-v0.1.1 and Equall/Saul-Instruct-v1. This merging approach aims to synthesize the capabilities of both base models into a single, more versatile instruction-following model.
Key Characteristics
- Merged Architecture: Utilizes a slerp merge method across specific layers (self_attn and mlp) to blend the characteristics of its two source models.
- Parameter Count: A 7 billion parameter model, offering a balance between performance and computational efficiency.
- Context Length: Supports a context window of 4096 tokens, enabling it to handle moderately long inputs and generate coherent responses.
Use Cases
This model is suitable for general instruction-following tasks where a merged model's combined strengths are beneficial. Its design suggests applicability in scenarios requiring robust natural language understanding and generation based on user prompts.