Join 📚 J'S Highlights
A batch of the best highlights from what J's read, .
In GPT-4, a word vector has thousands of dimensions, which describe its shades of similarity to and difference from every other word. During training, a large language model tweaks a word’s coördinates whenever it makes a prediction error; words that appear in texts together are nudged closer in space. This produces an incredibly dense representation of usages and meanings, in which analogy becomes a matter of geometry. In a classic example, if you take the word vector for “Paris,” subtract “France,” and then add “Italy,” the nearest other vector will be “Rome.”
The Case That A.I. Is Thinking | the New Yorker
James Somers
Brockman, who would later connect Epstein to dozens of world-famous scientists, most of them his clients, made it seem as if it were people like him who built this “third culture” – out of their perceptive genius. The cardinal error of such analysis, however, lies in its tendency to mistake structural transformations of global capitalism for zeitgeisty trends in the history of ideas.
The Epstein scandal at MIT shows the moral bankruptcy of techno-elites
Evgeny Morozov
What do I do now, for the next forty years? What work can’t the efficiency of unified mankind chop into pure fertilizer?
The Overstory
Richard Powers
...catch up on these, and many more highlights
Readwise