# Mckay Wrigley - Paul Graham GPT (Highlights)  ## Metadata **Review**:: [readwise.io](https://readwise.io/bookreview/24934327) **Source**:: #from/readwise **Zettel**:: #zettel/fleeting **Status**:: #x **Authors**:: [[Mckay Wrigley]] **Full Title**:: Paul Graham GPT **Category**:: #articles #readwise/articles **Category Icon**:: 📰 **URL**:: [github.com](https://github.com/mckaywrigley/paul-graham-gpt) **Host**:: [[github.com]] **Highlighted**:: [[2023-03-03]] **Created**:: [[2023-03-04]] ## Highlights ### Search - Search was created with [OpenAI Embeddings](https://platform.openai.com/docs/guides/embeddings) (`text-embedding-ada-002`). First, we loop over the essays and generate embeddings for each chunk of text. Then in the app we take the user's search query, generate an embedding, and use the result to find the most similar passages from the book. ([View Highlight](https://read.readwise.io/read/01gtkx3qdf9dqf5ec0xvwebk3a)) ^485470231 See the tool [gpt-index](https://github.com/jerryjliu/gpt_index) ### Chat - Chat builds on top of search. It uses search results to create a prompt that is fed into GPT-3.5-turbo. ([View Highlight](https://read.readwise.io/read/01gtkx47ev1smgsfgamn9p5zb4)) ^485470248 --- cover: "https://opengraph.githubassets.com/f1b9b714e12810a601d07bf5d9d3ace8dfd783a085e351eb9097b6d87228e1a4/mckaywrigley/paul-graham-gpt" --- # Mckay Wrigley - Paul Graham GPT (Highlights)  ## Metadata **Review**:: [readwise.io](https://readwise.io/bookreview/24934285) **Source**:: #from/readwise #from/reader **Zettel**:: #zettel/fleeting **Status**:: #x **Authors**:: [[Mckay Wrigley]] **Full Title**:: Paul Graham GPT **Category**:: #articles #readwise/articles **Category Icon**:: 📰 **URL**:: [github.com](https://github.com/mckaywrigley/paul-graham-gpt) **Host**:: [[github.com]] **Highlighted**:: [[2023-08-05]] **Created**:: [[2023-08-28]] ## Highlights ### Search - Search was created with [OpenAI Embeddings](https://platform.openai.com/docs/guides/embeddings) (`text-embedding-ada-002`). First, we loop over the essays and generate embeddings for each chunk of text. Then in the app we take the user's search query, generate an embedding, and use the result to find the most similar passages from the book. ([View Highlight](https://read.readwise.io/read/01gtkx3qdf9dqf5ec0xvwebk3a)) ^575278577 ### Chat - Chat builds on top of search. It uses search results to create a prompt that is fed into GPT-3.5-turbo. ([View Highlight](https://read.readwise.io/read/01gtkx47ev1smgsfgamn9p5zb4)) ^575278579