DataBricks Dolly 2.0, Industrial Use, TRUE Open Supply LLM

DataBricks Dolly 2.0, Industrial Use, TRUE Open Supply LLM

Databricks has launched a brand new language mannequin referred to as Dolly 2.0, which is an LLM with 12 billion parameters. The mannequin relies on the EleutherAI Pythia mannequin household and has been fine-tuned utilizing a brand new high-quality dataset of human-generated directions, which was crowdsourced amongst Databricks workers. This fine-tuning was achieved completely with an open-source dataset created by Databricks. Moreover, Databricks has made accessible the PyTorch code implementation of Dolly 2.0 on a free COLAB Pocket book.

Databricks is releasing all the Dolly 2.0 mannequin, together with the coaching code, dataset, and mannequin weights, for industrial use. Because of this any group can use, personal, and personalize highly effective LLMs that may talk with individuals, with out having to pay for API entry or share knowledge with third events. We specific our gratitude to Databricks for this initiative.

LINKS:Google Colab Pocket book: https://colab.research.google.com/drive/1iGnousCES7cxJXsDdLWWpPpfk9e9Ebzy?usp=sharingDolly 2.0 Weblog put up: https://www.databricks.com/blog/2023/04/12/dolly-first-open-commercially-viable-instruction-tuned-llmDolly 2.0 Fashions: https://huggingface.co/databricksDatabricks Information: https://github.com/databrickslabs/dolly/tree/master/dataEleutherAI: https://www.eleuther.ai/Pythia paper: https://arxiv.org/pdf/2304.01373.pdf

Purchase me a Espresso: https://ko-fi.com/promptengineeringBe part of the Patreon: patreon.com/PromptEngineering

#dolly #databricks #llm #chatgpt

The post DataBricks Dolly 2.0, Industrial Use, TRUE Open Supply LLM appeared first on AIPressRoom.