Weyaxi/Airboros2.1-Platypus2-13B-QLora-0.80-epoch
Weyaxi/Airboros2.1-Platypus2-13B-QLora-0.80-epoch is a 13 billion parameter language model based on the Platypus2 architecture, fine-tuned using QLoRA. This model demonstrates a balanced performance across various general language understanding and reasoning tasks, achieving an average score of 51.17 on the Open LLM Leaderboard. It is suitable for applications requiring general-purpose text generation and comprehension within a 4096-token context window.
Loading preview...
Airboros2.1-Platypus2-13B-QLora-0.80-epoch Overview
This model, developed by Weyaxi, is a 13 billion parameter language model built upon the Platypus2 architecture and fine-tuned with QLoRA. It is designed for general-purpose language tasks, operating within a 4096-token context window. The model's performance has been evaluated on the Open LLM Leaderboard, showcasing its capabilities across a range of benchmarks.
Key Performance Metrics
The model achieves an average score of 51.17 on the Open LLM Leaderboard, indicating its general proficiency. Specific benchmark results include:
- ARC (25-shot): 58.96
- HellaSwag (10-shot): 82.46
- MMLU (5-shot): 54.62
- TruthfulQA (0-shot): 47.71
- Winogrande (5-shot): 75.14
- DROP (3-shot): 39.32
Notably, the model scored 0.0 on GSM8K (5-shot), suggesting limitations in complex mathematical reasoning tasks. For detailed evaluation results, refer to the Open LLM Leaderboard evaluation page.
Ideal Use Cases
This model is well-suited for applications requiring:
- General text generation and completion.
- Understanding and responding to diverse prompts.
- Tasks where a balanced performance across various language understanding benchmarks is acceptable.