Myriad-7B-Slerp: A Merged 7B Language Model

Myriad-7B-Slerp is a 7 billion parameter language model developed by Muhammad2003, created through a Slerp merge of two distinct models: MaziyarPanahi/Calme-7B-Instruct-v0.9 and yam-peleg/Experiment26-7B. This merging technique, facilitated by LazyMergekit, aims to combine the capabilities of its base models to achieve enhanced performance.

Key Capabilities & Performance

This model operates with an 8192-token context window and has been evaluated across several common benchmarks, showcasing its general utility:

• ARC: 73.38
• Hellaswag: 89.05
• MMLU: 64.32
• TruthfulQA: 77.95
• Winogrande: 84.85
• GSM8k: 70.28

These scores indicate its proficiency in areas such as common sense reasoning, language understanding, and mathematical problem-solving.

When to Use This Model

Myriad-7B-Slerp is a strong candidate for applications requiring a capable 7B model with a balanced performance profile. Its merged architecture suggests a broad range of potential uses, including:

• General-purpose conversational AI
• Text generation and summarization
• Reasoning and question-answering tasks
• Educational tools and content creation