NekoPunchBBB/Llama-2-13b-hf_Open-Platypus-QLoRA-multigpu
NekoPunchBBB/Llama-2-13b-hf_Open-Platypus-QLoRA-multigpu is a 13 billion parameter language model based on the Llama 2 architecture. This model is fine-tuned using QLoRA and evaluated on the Open LLM Leaderboard, demonstrating a balanced performance across various benchmarks. It achieves an average score of 47.51, with notable results in ARC (57.51) and HellaSwag (82.49). This model is suitable for general-purpose language understanding and generation tasks, particularly where a 13B parameter model offers a good balance of performance and computational efficiency.
Loading preview...
Model Overview
NekoPunchBBB/Llama-2-13b-hf_Open-Platypus-QLoRA-multigpu is a 13 billion parameter language model built upon the Llama 2 architecture. It has been fine-tuned using the QLoRA method, which allows for efficient adaptation of large language models. The model's performance is officially evaluated on the Hugging Face Open LLM Leaderboard, providing transparent and comparable metrics.
Key Capabilities & Performance
This model demonstrates a solid, general-purpose understanding across a range of tasks, as indicated by its Open LLM Leaderboard scores. Key performance metrics include:
- Average Score: 47.51
- ARC (25-shot): 57.51
- HellaSwag (10-shot): 82.49
- MMLU (5-shot): 54.83
- TruthfulQA (0-shot): 43.81
- Winogrande (5-shot): 77.27
- GSM8K (5-shot): 10.46
- DROP (3-shot): 6.18
These scores suggest proficiency in common sense reasoning, reading comprehension, and general knowledge tasks, while indicating areas for potential improvement in complex mathematical reasoning (GSM8K) and factual accuracy (TruthfulQA).
Use Cases
Given its balanced performance, this model is well-suited for applications requiring:
- General text generation and completion
- Question answering based on provided context
- Summarization of short to medium-length texts
- Tasks benefiting from a 13B parameter model's efficiency and capability balance.