arlineka/Brunhilde-13b
Brunhilde-13b, developed by arlineka, is a 13 billion parameter language model with a 4096-token context window. It achieves an average score of 56.20 on the Open LLM Leaderboard, demonstrating capabilities across various benchmarks including MMLU (56.18) and HellaSwag (83.49). This model is suitable for general-purpose text generation and reasoning tasks, offering a balanced performance profile for its size.
Loading preview...
Brunhilde-13b Overview
Brunhilde-13b, developed by arlineka, is a 13 billion parameter language model designed for a variety of natural language processing tasks. It utilizes a 4096-token context window, making it suitable for processing moderately long inputs and generating coherent responses. The model's performance is publicly evaluated on the Open LLM Leaderboard, where it achieves a competitive average score of 56.20.
Key Capabilities
- General Text Generation: Capable of generating human-like text based on given prompts.
- Reasoning Tasks: Demonstrates proficiency in reasoning, as indicated by a 60.49 score on the AI2 Reasoning Challenge (25-Shot).
- Commonsense Inference: Achieves 83.49 on HellaSwag (10-Shot) and 75.53 on Winogrande (5-shot), suggesting strong commonsense understanding.
- Knowledge-based Question Answering: Scores 56.18 on MMLU (5-Shot), indicating a foundational understanding across various subjects.
Good for
- Prototyping and Development: A solid choice for developers looking for a 13B parameter model with balanced performance.
- General-purpose Chatbots: Its ability to handle diverse queries makes it suitable for conversational AI applications.
- Text Summarization and Generation: Can be used for tasks requiring the creation of new text or condensing existing information.
For detailed benchmark results, refer to the Open LLM Leaderboard evaluation page.