Model Overview

akjindal53244/Mistral-7B-v0.1-Open-Platypus is a 7 billion parameter language model built upon the Mistral-7B-v0.1 architecture. It has been instruction-finetuned using the comprehensive Open-Platypus dataset, enhancing its ability to follow instructions and perform a wide range of natural language tasks.

Key Capabilities & Performance

This model demonstrates strong performance across several benchmarks, as evaluated on the Hugging Face Open LLM Leaderboard. Its average score is 53.64, with notable results including:

• ARC (25-shot): 62.37
• HellaSwag (10-shot): 85.08
• MMLU (5-shot): 63.79
• Winogrande (5-shot): 77.66

While excelling in general reasoning and common sense, its performance on mathematical reasoning (GSM8K: 17.29) and reading comprehension (DROP: 21.93) indicates areas for potential improvement or specific use case considerations. The model supports a context length of 8192 tokens.

Good For

• General instruction-following tasks
• Applications requiring strong language understanding and generation
• Scenarios where a 7B parameter model with competitive benchmark performance is desired