Overview

jondurbin/airoboros-l2-13b-gpt4-1.4.1 is a 13 billion parameter language model built upon the Llama 2 architecture. It has been fine-tuned using a proprietary dataset, jondurbin/airoboros-gpt4-1.4.1, which was generated through OpenAI API calls to GPT-4. This training methodology aims to imbue the model with strong instruction-following and conversational abilities.

Key Characteristics

• Base Model: Llama 2 13B, providing a robust foundation for language understanding and generation.
• Training Data: Fine-tuned on a dataset derived from GPT-4 outputs, suggesting a focus on high-quality, diverse instruction-response pairs.
• Context Length: Supports a context window of 4096 tokens.

Licensing and Usage Considerations

This model inherits the proprietary/custom Meta license from its Llama 2 base. Additionally, the use of GPT-4 generated data for fine-tuning introduces complexities regarding OpenAI's Terms of Service, specifically concerning competitive use. Due to these licensing ambiguities and the origin of its training data, commercial use is advised against, and users assume full responsibility for compliance with all applicable licenses and terms.