Files
logseq/journals/2023_04_25.md
T
2025-06-02 17:15:13 +02:00

13 lines
2.1 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
- **22:54** [[quick capture]]:
La [[maratona di Londra]] che si è corsa domenica scorsa ha visto trionfare il 23enne keniano [[Kelvin Kiptum]]. che ha realizzato la seconda prestazione di tutti i tempi sulla distanza: con 2h01'25" è arrivato ad appena 16 secondi dal record mondiale che appartiene al connazionale Eliud Kipchoge, che lo aveva fatto segnare a Berlino lo scorso 25 settembre. Kiptum è stato velocissimo e si è preso gli applausi del folto pubblico che si è assiepato lungo il percorso della maratona inglese, ma un altro concorrente arrivato molto tempo dopo di lui ha dimostrato che il risultato non è tutto nello sport.
- type:: [[weblink]]
source:: [What Is ChatGPT Doing … and Why Does It Work?—Stephen Wolfram Writings](https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work/)
tags:: [[ReadingNotes]] [[chatgpt]]
date:: [[25-04-2023]] - 22:15
- There are about 40,000 [reasonably commonly used words in English](https://reference.wolfram.com/language/ref/WordList.html). And by looking at a large corpus of English text (say a few million books, with altogether a few hundred billion words), we can get an [estimate of how common each word is](https://reference.wolfram.com/language/ref/WordFrequencyData.html).
- using this we can start generating “sentences”, in which each word is independently picked at random, with the same probability that it appears in the corpus.
- we can start taking into account not just probabilities for single words but probabilities for pairs or longer __n__\-grams of words
- In a [crawl of the web](https://commoncrawl.org/) there might be a few hundred billion words; in how long its going to take a cannon ball dropped from each floor of the Tower of Pisa to hit the ground.
- make a model that gives some kind of procedure for computing the answer rather than just measuring and remembering each case
- Any model you use has some particular underlying structure—then a certain set of “knobs you can turn” (i.e. parameters you can set) to fit your data.
- 175 billion of them