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.

roman01la

#!/usr/bin/env bash
set -euo pipefail

# patch-claude-code.sh — Rebalance Claude Code prompts to fix corner-cutting behavior
#
# What this does:
#   Patches the npm-installed @anthropic-ai/claude-code cli.js to rebalance
#   system prompt instructions that cause the model to cut corners, simplify
#   excessively, and defer complicated work.
#

rcdilorenzo

const puppeteer = require('puppeteer');
const read = require('read');
const htmlToText = require('html-to-text').fromString;
const R = require('ramda');
const Promise = require('bluebird');
const fs = require('fs');
const download = require('@jinphen/download2');
const { CookieJar } = require('tough-cookie');

const mapSeries = R.flip(Promise.mapSeries);

abalter

• Install EVERYTHING you can with conda
• Install things you can with:
  • install.packages(, lib="")
  • install.github(, lib="")
  • biocLite(, lib="")
• Make separate local libs for different R versions
• After installing a package to a local lib, look there and see if there are dependencies you could have installed with conda.
• Or, head it off by looking at the dependencies in CRAN or Bioconductor

talegari

###############################################################################
#
# cor2
# -- Compute correlations of columns of a dataframe of mixed types
#
###############################################################################
#
# author  : Srikanth KS (talegari)
# license : GNU AGPLv3 (http://choosealicense.com/licenses/agpl-3.0/)
#

mjdietzx

#!/bin/bash

# install CUDA Toolkit v8.0
# instructions from https://developer.nvidia.com/cuda-downloads (linux -> x86_64 -> Ubuntu -> 16.04 -> deb (network))
CUDA_REPO_PKG="cuda-repo-ubuntu1604_8.0.61-1_amd64.deb"
wget http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/${CUDA_REPO_PKG}
sudo dpkg -i ${CUDA_REPO_PKG}
sudo apt-get update
sudo apt-get -y install cuda

ksopyla

# This is shorthened version of blog post 
#  http://ksopyla.com/2017/02/tensorflow-gpu-virtualenv-python3/

# update packages
sudo apt-get update
sudo apt-get upgrade

#Add the ppa repo for NVIDIA graphics driver
sudo add-apt-repository ppa:graphics-drivers/ppa
sudo apt-get update

binary1230

3/18/2021: WE HAVE MOVED: For the latest instructions on how to use the bulk wells fargo PDF downloader, 
please ignore this page and visit below:

https://github.com/binary1230/wellsfargo-bulk-PDF-statement-downloader/blob/main/README.md

awjuliani

mikeboers

import argparse
import datetime
import json
import os
import re

import requests


parser = argparse.ArgumentParser()