42MARU/GenAI-llama-2-13b
42MARU/GenAI-llama-2-13b is a 13 billion parameter language model developed by 42MARU, built upon the LLaMA-2 backbone. It is fine-tuned on Orca-style and Platypus datasets, making it suitable for general conversational AI and question-answering tasks. This model demonstrates solid performance across various benchmarks, including ARC, HellaSwag, and MMLU, indicating its utility for diverse natural language understanding applications.
Loading preview...
Model Overview
42MARU/GenAI-llama-2-13b is a 13 billion parameter language model developed by 42MARU, leveraging the LLaMA-2 architecture. It has been fine-tuned using Orca-style and Platypus datasets, enhancing its capabilities for conversational AI and question-answering.
Key Capabilities & Performance
This model demonstrates competitive performance across several benchmarks on the Open LLM Leaderboard:
- Avg. Score: 56.03
- ARC (25-shot): 63.14
- HellaSwag (10-shot): 83.64
- MMLU (5-shot): 59.91
- TruthfulQA (0-shot): 56.21
- Winogrande (5-shot): 76.72
- GSM8K (5-shot): 9.4
- DROP (3-shot): 43.23
These scores indicate its proficiency in common sense reasoning, reading comprehension, and general knowledge tasks. The model utilizes a standard prompt template for user and assistant interactions.
Use Cases
Given its fine-tuning and benchmark performance, 42MARU/GenAI-llama-2-13b is well-suited for:
- General-purpose conversational agents: Engaging in natural dialogue.
- Question Answering systems: Providing informative responses based on context.
- Text generation tasks: Creating coherent and relevant text.
- Educational applications: Assisting with understanding and knowledge retrieval.