abacusai/MM-Orc-Vic-bagel-34b-c1000
TEXT GENERATIONConcurrency Cost:2Model Size:34BQuant:FP8Ctx Length:32kPublished:Jan 19, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold
The abacusai/MM-Orc-Vic-bagel-34b-c1000 is a 34 billion parameter language model, fine-tuned from the nontoxic-bagel-34b-v0.2 architecture. This model specializes in mathematical reasoning, having been trained on the MetaMathFewshot dataset. It demonstrates enhanced performance on mathematical benchmarks like GSM8K compared to its base model, making it suitable for complex quantitative tasks.
Loading preview...
Model Overview
The abacusai/MM-Orc-Vic-bagel-34b-c1000 is a 34 billion parameter language model, developed by abacusai. It is a fine-tuned version of the jondurbin/nontoxic-bagel-34b-v0.2 model, specifically optimized for mathematical reasoning tasks.
Key Capabilities
- Enhanced Mathematical Reasoning: The model has been fine-tuned on the MetaMathFewshot dataset, which focuses on improving mathematical problem-solving abilities.
- Improved GSM8K Performance: Evaluation results indicate a significant improvement in the GSM8K benchmark score compared to its base model. The
MM-Orc-Vic-bagel-34b-c1000achieves a higher GSM8K score than the originalnontoxic-bagel-34b-v0.2model's 58.45.
Good For
- Mathematical Problem Solving: Ideal for applications requiring strong quantitative reasoning and accurate solutions to mathematical problems.
- Research and Development: Suitable for researchers exploring advanced mathematical capabilities in large language models.
- Educational Tools: Can be integrated into tools designed to assist with learning or solving complex math problems.