Weyaxi/Athena-Platypus2-13B-QLora-0.80-epoch
Weyaxi/Athena-Platypus2-13B-QLora-0.80-epoch is a 13 billion parameter language model, fine-tuned using QLoRA, achieving an average score of 48.78 on the Open LLM Leaderboard. This model demonstrates capabilities across various reasoning and common sense tasks, including ARC, HellaSwag, MMLU, and Winogrande. It is suitable for applications requiring general language understanding and generation, with specific performance metrics available for detailed evaluation.
Loading preview...
Model Overview
Weyaxi/Athena-Platypus2-13B-QLora-0.80-epoch is a 13 billion parameter language model that has undergone QLoRA fine-tuning. This model's performance is evaluated on the Open LLM Leaderboard, where it achieved an average score of 48.78.
Key Capabilities & Performance
The model demonstrates varying levels of proficiency across several benchmarks:
- ARC (25-shot): 56.66
- HellaSwag (10-shot): 80.56
- MMLU (5-shot): 55.43
- TruthfulQA (0-shot): 53.62
- Winogrande (5-shot): 72.61
- GSM8K (5-shot): 0.08
- DROP (3-shot): 22.51
These scores indicate its strengths in common sense reasoning (HellaSwag, Winogrande) and general knowledge (MMLU), while showing lower performance in complex mathematical reasoning (GSM8K) and reading comprehension (DROP). Detailed evaluation results are available on the Hugging Face Open LLM Leaderboard.
Use Cases
This model is suitable for applications that benefit from its demonstrated capabilities in:
- General text generation and understanding
- Common sense reasoning tasks
- Question answering where factual recall and logical inference are key.
Developers should consider the specific benchmark scores when evaluating its fit for tasks requiring high accuracy in areas like complex math or deep reading comprehension.