Matches in SemOpenAlex for { <https://semopenalex.org/work/W4315706637> ?p ?o ?g. }
- W4315706637 abstract "The BigCode project is an open-scientific collaboration working on the responsible development of large language models for code. This tech report describes the progress of the collaboration until December 2022, outlining the current state of the Personally Identifiable Information (PII) redaction pipeline, the experiments conducted to de-risk the model architecture, and the experiments investigating better preprocessing methods for the training data. We train 1.1B parameter models on the Java, JavaScript, and Python subsets of The Stack and evaluate them on the MultiPL-E text-to-code benchmark. We find that more aggressive filtering of near-duplicates can further boost performance and, surprisingly, that selecting files from repositories with 5+ GitHub stars deteriorates performance significantly. Our best model outperforms previous open-source multilingual code generation models (InCoder-6.7B and CodeGen-Multi-2.7B) in both left-to-right generation and infilling on the Java, JavaScript, and Python portions of MultiPL-E, despite being a substantially smaller model. All models are released under an OpenRAIL license at https://hf.co/bigcode." @default.
- W4315706637 created "2023-01-12" @default.
- W4315706637 creator A5000043237 @default.
- W4315706637 creator A5000126238 @default.
- W4315706637 creator A5000510528 @default.
- W4315706637 creator A5001782949 @default.
- W4315706637 creator A5002251783 @default.
- W4315706637 creator A5003637850 @default.
- W4315706637 creator A5005743381 @default.
- W4315706637 creator A5007842855 @default.
- W4315706637 creator A5008355834 @default.
- W4315706637 creator A5009823475 @default.
- W4315706637 creator A5010465328 @default.
- W4315706637 creator A5018410563 @default.
- W4315706637 creator A5018675797 @default.
- W4315706637 creator A5029389486 @default.
- W4315706637 creator A5033794353 @default.
- W4315706637 creator A5035649010 @default.
- W4315706637 creator A5044042294 @default.
- W4315706637 creator A5044824055 @default.
- W4315706637 creator A5045436969 @default.
- W4315706637 creator A5048579807 @default.
- W4315706637 creator A5054634361 @default.
- W4315706637 creator A5054948010 @default.
- W4315706637 creator A5055115466 @default.
- W4315706637 creator A5063189132 @default.
- W4315706637 creator A5068788224 @default.
- W4315706637 creator A5070647299 @default.
- W4315706637 creator A5071825886 @default.
- W4315706637 creator A5072906919 @default.
- W4315706637 creator A5075446366 @default.
- W4315706637 creator A5075492610 @default.
- W4315706637 creator A5076596867 @default.
- W4315706637 creator A5077190748 @default.
- W4315706637 creator A5077804559 @default.
- W4315706637 creator A5082114795 @default.
- W4315706637 creator A5082624117 @default.
- W4315706637 creator A5083384339 @default.
- W4315706637 creator A5084438979 @default.
- W4315706637 creator A5085873515 @default.
- W4315706637 creator A5088568785 @default.
- W4315706637 creator A5090893148 @default.
- W4315706637 creator A5091564502 @default.
- W4315706637 date "2023-01-09" @default.
- W4315706637 modified "2023-10-14" @default.
- W4315706637 title "SantaCoder: don't reach for the stars!" @default.
- W4315706637 doi "https://doi.org/10.48550/arxiv.2301.03988" @default.
- W4315706637 hasPublicationYear "2023" @default.
- W4315706637 type Work @default.
- W4315706637 citedByCount "0" @default.
- W4315706637 crossrefType "posted-content" @default.
- W4315706637 hasAuthorship W4315706637A5000043237 @default.
- W4315706637 hasAuthorship W4315706637A5000126238 @default.
- W4315706637 hasAuthorship W4315706637A5000510528 @default.
- W4315706637 hasAuthorship W4315706637A5001782949 @default.
- W4315706637 hasAuthorship W4315706637A5002251783 @default.
- W4315706637 hasAuthorship W4315706637A5003637850 @default.
- W4315706637 hasAuthorship W4315706637A5005743381 @default.
- W4315706637 hasAuthorship W4315706637A5007842855 @default.
- W4315706637 hasAuthorship W4315706637A5008355834 @default.
- W4315706637 hasAuthorship W4315706637A5009823475 @default.
- W4315706637 hasAuthorship W4315706637A5010465328 @default.
- W4315706637 hasAuthorship W4315706637A5018410563 @default.
- W4315706637 hasAuthorship W4315706637A5018675797 @default.
- W4315706637 hasAuthorship W4315706637A5029389486 @default.
- W4315706637 hasAuthorship W4315706637A5033794353 @default.
- W4315706637 hasAuthorship W4315706637A5035649010 @default.
- W4315706637 hasAuthorship W4315706637A5044042294 @default.
- W4315706637 hasAuthorship W4315706637A5044824055 @default.
- W4315706637 hasAuthorship W4315706637A5045436969 @default.
- W4315706637 hasAuthorship W4315706637A5048579807 @default.
- W4315706637 hasAuthorship W4315706637A5054634361 @default.
- W4315706637 hasAuthorship W4315706637A5054948010 @default.
- W4315706637 hasAuthorship W4315706637A5055115466 @default.
- W4315706637 hasAuthorship W4315706637A5063189132 @default.
- W4315706637 hasAuthorship W4315706637A5068788224 @default.
- W4315706637 hasAuthorship W4315706637A5070647299 @default.
- W4315706637 hasAuthorship W4315706637A5071825886 @default.
- W4315706637 hasAuthorship W4315706637A5072906919 @default.
- W4315706637 hasAuthorship W4315706637A5075446366 @default.
- W4315706637 hasAuthorship W4315706637A5075492610 @default.
- W4315706637 hasAuthorship W4315706637A5076596867 @default.
- W4315706637 hasAuthorship W4315706637A5077190748 @default.
- W4315706637 hasAuthorship W4315706637A5077804559 @default.
- W4315706637 hasAuthorship W4315706637A5082114795 @default.
- W4315706637 hasAuthorship W4315706637A5082624117 @default.
- W4315706637 hasAuthorship W4315706637A5083384339 @default.
- W4315706637 hasAuthorship W4315706637A5084438979 @default.
- W4315706637 hasAuthorship W4315706637A5085873515 @default.
- W4315706637 hasAuthorship W4315706637A5088568785 @default.
- W4315706637 hasAuthorship W4315706637A5090893148 @default.
- W4315706637 hasAuthorship W4315706637A5091564502 @default.
- W4315706637 hasBestOaLocation W43157066371 @default.
- W4315706637 hasConcept C118643609 @default.
- W4315706637 hasConcept C136764020 @default.
- W4315706637 hasConcept C137293760 @default.
- W4315706637 hasConcept C154945302 @default.
- W4315706637 hasConcept C199360897 @default.
- W4315706637 hasConcept C2777904410 @default.
- W4315706637 hasConcept C3018397939 @default.