Languages used to train BLOOM BLOOM has been trained on thirteen languages in the Indic family from the Indian subcontinent (e.g., Hindi, Tamil, and Urdu) and twenty sub-Saharan languages in the Niger-Congo family (e.g., Swahili, Yoruba, and Wolof). Code in thirteen different programming languages accounted for 10.8% of its input. Source: Hugging Face
Languages used to train BLOOM BLOOM has been trained on thirteen languages in the Indic family from the Indian subcontinent (e.g. Hindi, Tamil, and Urdu) and twenty sub-Saharan languages in the Niger-Congo family (e.g. Swahili, Yoruba, and Wolof). Code in thirteen different programming languages accounted for 10.8% of its input. Source: Hugging Face - Language models are AI systems whose primary applications concern natural (i.e., human) language. They might answer questions; generate sentences; detect emotions; or summarize, simplify, or translate text. Usually designed by giant tech firms, most existing models have solely been trained with English text and apply principles and methods that are difficult to fully replicate.
TO READ THIS ARTICLE, CREATE YOUR ACCOUNT
And extend your reading, free of charge and with no commitment.
