Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model.
Shaden SmithMostofa PatwaryBrandon NorickPatrick LeGresleySamyam RajbhandariJared CasperZhun LiuShrimai PrabhumoyeGeorge ZerveasVijay KorthikantiElton ZhengRewon ChildReza Yazdani AminabadiJulie BernauerXia SongMohammad ShoeybiYuxiong HeMichael HoustonSaurabh TiwaryBryan CatanzaroPublished in: CoRR (2022)
Keyphrases
- language model
- language modeling
- probabilistic model
- document retrieval
- speech recognition
- information retrieval
- language modelling
- generative model
- query expansion
- retrieval model
- n gram
- context sensitive
- mixture model
- ad hoc information retrieval
- query terms
- vector space model
- pseudo relevance feedback
- test collection
- language modeling framework
- statistical language models
- smoothing methods
- document ranking
- document length
- natural language
- translation model
- relevance model
- language model for information retrieval
- unsupervised learning
- machine learning
- word clouds