Dolphin

Today I am announcing Dolphin, an open-source and uncensored, and commercially licensed dataset and series of instruct-tuned language models based on Microsoft's Orca paper. The dataset is released here, with Apache-2.0 license. The dataset can be us...