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.

WardBrian

from jax import random, jit
import jax.numpy as jnp
from jax.scipy import stats

from util import ravelize_function, make_log_density

__all__ = ["log_density", "log_density_vec", "init_draw_zero"]


def constrain_parameters(sigma_unc, alpha, beta):

TheMatt2

Checkbox Tables in Markdown

An attempt to make a list of the supported ways to make a table with checkboxes in Markdown.

Results as of October 2023.

---

Below is the style element that formats the colors of the colored check mark emojis.

peterhurford

brew install pandoc
brew tap homebrew/cask
brew install --cask basictex
eval "$(/usr/libexec/path_helper)"
# Update $PATH to include `/usr/local/texlive/2022basic/bin/universal-darwin`
sudo tlmgr update --self
sudo tlmgr install texliveonfly
sudo tlmgr install xelatex
sudo tlmgr install adjustbox
sudo tlmgr install tcolorbox