Dolly 2.0 7B

About

Dolly 2.0 7B is an instruction-following large language model developed by Databricks. It is based on EleutherAI's Pythia-6.9b and fine-tuned using a dataset of around 15,000 instruction-response pairs created by Databricks employees. This model is distinguished by its open-source nature and permissive license, supporting both research and commercial use. Although not state-of-the-art, Dolly 2.0 7B is proficient in tasks like brainstorming, classification, question answering, text generation, and summarization. It faces challenges with complex prompts, math operations, and factual accuracy. Application areas include AI assistants, content creation, and data analysis, and it is available in larger (12B) and smaller (3B) parameter versions.

Capabilities