Hugging Face MaziyarPanahi/calme-2.3-qwen2-7b is a 7.6 billion parameter causal language model, fine-tuned by MaziyarPanahi from the Qwen/Qwen2-7B architecture. This model aims to enhance the base Qwen2-7B performance across various benchmarks, demonstrating an average score of 22.74 on the Open LLM Leaderboard. With a substantial context length of 131072 tokens, it is optimized for general-purpose language tasks and improved reasoning capabilities.
Model Overview
MaziyarPanahi/calme-2.3-qwen2-7b is a 7.6 billion parameter language model developed by MaziyarPanahi. It is a fine-tuned iteration of the Qwen/Qwen2-7B base model, specifically designed to achieve improved performance across a range of benchmarks.
Key Capabilities & Performance
This model demonstrates enhanced capabilities as reflected in its Open LLM Leaderboard evaluation results. Key performance metrics include:
- Average Score: 22.74
- IFEval (0-Shot): 38.25
- BBH (3-Shot): 30.96
- MATH Lvl 5 (4-Shot): 18.66
- MMLU-PRO (5-shot): 29.01
The model utilizes the ChatML prompt template, making it compatible with standard instruction-following formats. It also supports a large context window of 131072 tokens, allowing for processing extensive inputs.
Usage Considerations
- Fine-tuned for general improvement: This model is intended for users seeking a more capable version of the Qwen2-7B base model for various language generation and understanding tasks.
- Quantized versions available: For users requiring optimized inference, quantized GGUF models are provided at MaziyarPanahi/calme-2.3-qwen2-7b-GGUF.
- Prompt format: Adheres to the
ChatMLformat for structured conversations.