Will Koehrsen WillKoehrsen

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

"POOR BILL"
2019-Nov-07

@teddyschleifer wrote:

"Bill Gates on a wealth tax:

'I've paid over $10 billion in taxes. I've paid more than anyone in taxes. If I had to pay $20 billion, it's fine.'

'But when you say I should pay $100 billion, then I'm starting to do a little math over what I have left over.'"

How to add a collapsible section in markdown

1. Example

Click me

Heading

Foo
Bar
- Baz

Qux

Using py.test is great and the support for test fixtures is pretty awesome. However, in order to share your fixtures across your entire module, py.test suggests you define all your fixtures within one single conftest.py file. This is impractical if you have a large quantity of fixtures -- for better organization and readibility, you would much rather define your fixtures across multiple, well-named files. But how do you do that? ...No one on the internet seemed to know.

Turns out, however, you can define fixtures in individual files like this:

tests/fixtures/add.py

import pytest

@pytest.fixture

	#!/bin/bash

	echo 'update brew'

	brew update

	echo 'upgrade brew'

	brew upgrade

	# for spyshell
	function zshexit() {
	cat /Users/`whoami`/.termlogs/`date +%Y-%m-%d`.txt \| perl -pe 's/\e([^\[\]]\|\[.?[a-zA-Z]\|\].?\a)//g' \| col -b > temp
	mv temp /Users/`whoami`/.termlogs/`date +%Y-%m-%d`.txt
	exit
	}