karpathy

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

karpathy

"""
The most atomic way to train and run inference for a GPT in pure, dependency-free Python.
This file is the complete algorithm.
Everything else is just efficiency.

@karpathy
"""

import os       # os.path.exists
import math     # math.log, math.exp

JoshuaGross

Freeing up extra disk space inside a Docker container

Sometimes while running a command inside of a docker container, the container itself will run out of disk space and things can fail (sometimes in strange ways). This is not a guide on how to allocate resources, but a quick cookbook on how to free up disk space inside a container. See below for links to more resources and reading.

If at all possible, just run Docker with increased disk space

See Limit disk size and bandwidth of a Docker container.

Resize an aufs volume

arkihillel

<dom-module id="webcam-capture">
	<template>
		<style>
			:host {
				display: block;
				top: 0;
				bottom: 0;
				left: 0;
				right: 0;
				position: fixed;

robulouski

#!/usr/bin/env python
#
# Very basic example of using Python and IMAP to iterate over emails in a
# gmail folder/label.  This code is released into the public domain.
#
# RKI July 2013
# http://www.voidynullness.net/blog/2013/07/25/gmail-email-with-python-via-imap/
#
import sys
import imaplib

splorp

body {
	background-color: #FFF;
	color: #333;
	font: 15px Helvetica, arial, freesans, clean, sans-serif;
	word-wrap: break-word;
	line-height: 1.7;
	padding: 0 20px 20px 20px;
	width: 722px;
	-webkit-font-smoothing: antialiased;
	}