New Scientist article on LLMs 2023

Activity Type:

Media interview, press release, press conference or other response to a media enquiry

How many people:

More than 500

Geographical Reach:

International

Primary Audience:

Media (as a channel to wider audiences)

Other Audience:

General public

Activity Years:

2023

Result Description:

Interview about pruning large language models based on my expertise from participating in the BigScience BLOOM modelling collaboration: "Huge AI models can be halved in size without degrading performance A way to cut the scale of artificial intelligence models by 60 per cent could save huge amounts of energy and make them more accessible"

Most important impact:

Not aware of any impact

URL:

https://www.newscientist.com/article/2355078-huge-ai-models-can-be-halved-in-size-without-degrading-performance/

Digital ID:

Source:

Manual

Publication ID:

63f39924708333.63355637