Weyaxi/Einstein-v2-7B
Weyaxi/Einstein-v2-7B is a 7 billion parameter causal language model developed by Weyaxi, fine-tuned from Mistral-7B-v0.1. This model is optimized for general reasoning and language understanding, achieving an average score of 63.48 on the Open LLM Leaderboard across various benchmarks including MMLU and HellaSwag. It is suitable for a range of natural language processing tasks requiring robust comprehension and generation capabilities.
Loading preview...
Weyaxi/Einstein-v2-7B: A Fine-Tuned Mistral Model
Weyaxi/Einstein-v2-7B is a 7 billion parameter language model developed by Weyaxi, building upon the mistralai/Mistral-7B-v0.1 architecture. This iteration, version 2, has been fine-tuned using the Axolotl framework, indicating a focus on efficient and structured training methodologies.
Key Capabilities & Performance
The model demonstrates solid performance across a suite of general language understanding and reasoning benchmarks, as evaluated on the Open LLM Leaderboard. It achieves an average score of 63.48, with notable results including:
- AI2 Reasoning Challenge (25-Shot): 62.37
- HellaSwag (10-Shot): 83.46
- MMLU (5-Shot): 62.08
- Winogrande (5-Shot): 79.32
These scores suggest proficiency in common sense reasoning, multiple-choice question answering, and general knowledge tasks. The model was trained with a learning rate of 5e-06 over one epoch, utilizing a sequence length of 8192 tokens and flash attention for efficiency.
Intended Use Cases
Given its foundational Mistral architecture and fine-tuning, Weyaxi/Einstein-v2-7B is well-suited for a variety of applications requiring strong language comprehension and generation. Developers can leverage this model for tasks such as:
- General-purpose chatbots and conversational AI.
- Text summarization and content generation.
- Reasoning tasks and question answering.
- Applications benefiting from a balanced performance across diverse benchmarks.