A jargon-free explanation of how AI large language models work
arstechnica.com
LLMs like ChatGPT are able to represent the same word with different vectors depending on the context in which that word appears. There’s a vector for bank (financial institution) and a different vector for bank (of a river). There’s a vector for magazine (physical publication) and another for magaz
2 Users
0 Comments
8 Highlights
6 Notes
Top Highlights
- LLMs like ChatGPT are able to represent the same word with different vectors depending on the context in which that word appears. There’s a vector for bank (financial institution) and a different vector for bank (of a river). There’s a vector for magazine (physical publication) and another for magazine (organization). As you might expect, LLMs use ...
- ChatGPT are able to represent the same word with different vectors depending on the context in which that word appears.
- Google announced its word2vec project
- LLMs use more similar vectors for polysemous meanings than homonymous ones
Domain
Ready to highlight and find good content?
Glasp is a social web highlighter that people can highlight and organize quotes and thoughts from the web, and access other like-minded people’s learning.