We present H2O-Danube-1.8B, a 1.8B language model trained on 1T tokens following the core principles of LLama 2 and Mistral. We leverage and refine various techniques for pre-training large language models. Although our model is trained on significantly fewer total tokens compared to reference models of similar size, it exhibits highly competitive metrics across a multitude of benchmarks. We additionally release a chat model trained with supervised fine-tuning followed by direct preference optimization. We make H2O-Danube-1.8B openly available under Apache 2.0 license further democratizing LLMs to a wider audience economically.Base model: https://huggingface.co/h2oai/h2o-danube-1.8b-base Chat model: https://huggingface.co/h2oai/h2o-danube-1.8b-chat * The first two authors contributed equally.
If, to misquote Harold Lasswell, one considers the study of politics to be the study of who gets what, when, and why, then this paper is a study of the politics of elections in the United Nations. Who gets what and when are easily discovered, since the results of elections and dates of elections are available in any United Nations Yearbook. The “why” is more difficult to determine. This paper is an attempt to analyze—by the use of empirical, numerical indices exclusively—why nations are elected to UN offices.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.