Pham Thanh Lam lampts

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

ChatGPT Resources

Context

ChatGPT appeared like an explosion on all my social media timelines in early December 2022. While I keep up with machine learning as an industry, I wasn't focused so much on this particular corner, and all the screenshots seemed like they came out of nowhere. What was this model? How did the chat prompting work? What was the context of OpenAI doing this work and collecting my prompts for training data?

I decided to do a quick investigation. Here's all the information I've found so far. I'm aggregating and synthesizing it as I go, so it's currently changing pretty frequently.

Model Architecture

Setting up postfix on MacOS High Sierra

MacOS Sierra High comes with preinstalled postfix. Follow these steps to configure it on a local system:

1. Create a `sasl_passwd` file

sudo nano /etc/postfix/sasl_passwd

Add this to the file

I use PlantUML a lot. It's what I use for drawing all sorts of diagrams and it's handy because of its easy markup (once you get used to it) while making things easy to maintain as projects grow (thanks to version control)

This gist explains how I do my PlantUML workspace in a project.

The idea is to keep a globals directory for all diagrams to follow (like the "stylesheet" below) to keep things consistent.
I use a stylesheet.iuml file that keeps the use of colors consistent through use of basic FOREGROUND, BACKGROUND and ACCENT colors.
The style-presets.iuml file defines these colors so you can make "presets" or "themes" out of them.
As stated in the stylesheet.iuml, you'll need the Roboto Condensed and Inconsolata fonts for these to work properly.
You can choose to either run the PlantUML jar over your file/s, or use an IDE like VSCode with the PlantUML extension. Here's a preview of example-sequence.puml for example: https://imgur.com/Klk3w2F

	# tutorial video link : https://youtu.be/dYt9xJ7dnpU
	# colab link : https://colab.research.google.com/drive/1xSbu-b-EwYd6GdaFPRVgvXBX_mciZ41e?usp=sharing
	# repo link : https://github.com/ai-forever/Kandinsky-2
	# used repo commit hash : a4354c04d5fbd48851866ef7d84ec444d3d50102

	# those who getting cuda error
	# pip uninstall torch
	# pip3 install torch==1.13.1 torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117

	import os

	import os
	os.environ["OPENAI_API_KEY"] = ""

	from flask import Flask, Response, request
	import threading
	import queue

	from langchain.chat_models import ChatOpenAI
	from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
	from langchain.schema import AIMessage, HumanMessage, SystemMessage

	from typing import TypeVar, Generic, Callable
	from dataclasses import dataclass
	from argparse import Namespace

	T = TypeVar('T')
	S = TypeVar('S')

	@dataclass
	class ListMap(Generic[S, T]):
	f: Callable[[T], S]

	FROM registry.visionhub.ru/models/base:v5 AS base

	# ============ BEGIN User model environment ============
	FROM ubuntu:18.04

	RUN apt-get update && apt-get install -y libsm6 libxext6 libxrender-dev libglib2.0-0 ffmpeg zlib1g-dev libjpeg-dev
	RUN apt-get install -y python3-pip
	RUN update-alternatives --install /usr/bin/python python /usr/bin/python3 1
	RUN update-alternatives --install /usr/bin/pip pip /usr/bin/pip3 1
	RUN pip install -U pip

	""" Use Apple's Vision Framework via PyObjC to detect text in images

	To use:

	python3 -m pip install pyobjc-core pyobjc-framework-Quartz pyobjc-framework-Vision wurlitzer

	"""

	import pathlib

	import argparse
	import math
	import os
	from multiprocessing import Pool, cpu_count

	import numpy as np
	import pandas as pd
	import tensorflow as tf
	from tqdm import tqdm