Quan Dang 18520339

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.

Reinforcement Learning for Language Models

Yoav Goldberg, April 2023.

Why RL?

With the release of the ChatGPT model and followup large language models (LLMs), there was a lot of discussion of the importance of "RLHF training", that is, "reinforcement learning from human feedback". I was puzzled for a while as to why RL (Reinforcement Learning) is better than learning from demonstrations (a.k.a supervised learning) for training language models. Shouldn't learning from demonstrations (or, in language model terminology "instruction fine tuning", learning to immitate human written answers) be sufficient? I came up with a theoretical argument that was somewhat convincing. But I came to realize there is an additional argumment which not only supports the case of RL training, but also requires it, in particular for models like ChatGPT. This additional argument is spelled out in (the first half of) a talk by John Schulman from OpenAI. This post pretty much

Chat GPT "DAN" (and other "Jailbreaks")

Clone the repo.
Use git rebase -i --root
vim will open. Select the commits you want to modify by changing pick to edit. If you would like to change all the commits, perform the following replace: :%s/^pick/edit/g. This command changes all instances of "pick" at the start of lines to "edit".
You will now be shown all the selected commits one by one. Each commit message will be displayed. You have two options:
- If you would like to keep the commit author details the same, do a git rebase --continue.
- If you would like to change it to a different name/email, do git commit --amend --reset-author. If --reset-author is specified, it will use the details from your git config. (If you need to specify an alternate name/email, you can do so with --author="John Doe <john@example.com>". If you would like to change the time to a previous date, you can do so with --date "2 days ago".)
Do the same for all the commits and finish the rebase.
Perform git push -f origin master to

	"""
	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

	# -----------------------------------------------------------------------------
	# AI-powered Git Commit Function
	# Copy paste this gist into your ~/.bashrc or ~/.zshrc to gain the `gcm` command. It:
	# 1) gets the current staged changed diff
	# 2) sends them to an LLM to write the git commit message
	# 3) allows you to easily accept, edit, regenerate, cancel
	# But - just read and edit the code however you like
	# the `llm` CLI util is awesome, can get it here: https://llm.datasette.io/en/stable/

	gcm() {

	const fs = require('fs');
	const axios = require('axios');
	const FormData = require('form-data');

	// Function to read the image file as a buffer
	const readImageFile = (filePath) => {
	return new Promise((resolve, reject) => {
	fs.readFile(filePath, (err, data) => {
	if (err) {
	reject(err);

	import torch
	import numpy as np
	from typing import List
	import ctc_segmentation
	from datasets import load_dataset
	from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC, Wav2Vec2CTCTokenizer

	# load model, processor and tokenizer
	model_name = "jonatasgrosman/wav2vec2-large-xlsr-53-english"
	processor = Wav2Vec2Processor.from_pretrained(model_name)

	#######################################
	### -------- Load libraries ------- ###

	# Load Huggingface transformers
	from transformers import TFBertModel, BertConfig, BertTokenizerFast

	# Then what you need from tensorflow.keras
	from tensorflow.keras.layers import Input, Dropout, Dense
	from tensorflow.keras.models import Model
	from tensorflow.keras.optimizers import Adam

	# -- coding: utf-8 --

	import regex as re

	uniChars = "àáảãạâầấẩẫậăằắẳẵặèéẻẽẹêềếểễệđìíỉĩịòóỏõọôồốổỗộơờớởỡợùúủũụưừứửữựỳýỷỹỵÀÁẢÃẠÂẦẤẨẪẬĂẰẮẲẴẶÈÉẺẼẸÊỀẾỂỄỆĐÌÍỈĨỊÒÓỎÕỌÔỒỐỔỖỘƠỜỚỞỠỢÙÚỦŨỤƯỪỨỬỮỰỲÝỶỸỴÂĂĐÔƠƯ"
	unsignChars = "aaaaaaaaaaaaaaaaaeeeeeeeeeeediiiiiooooooooooooooooouuuuuuuuuuuyyyyyAAAAAAAAAAAAAAAAAEEEEEEEEEEEDIIIOOOOOOOOOOOOOOOOOOOUUUUUUUUUUUYYYYYAADOOU"


	def loaddicchar():
	dic = {}