Peter Cooper peterc

A test of the technique illustrated by Simon Willison at https://til.simonwillison.net/llms/docs-from-tests

JSON Library Usage Documentation

Basic Usage

Parsing JSON

# Parse JSON string into Ruby objects

To compile:

as -o hello.o hello.s
ld -o hello hello.o -lSystem -syslibroot `xcrun -sdk macosx --show-sdk-path` -e _main -arch arm64

./hello

A 404 page.. kinda

The content which was previous linked to by the link you clicked has been removed for quality control reasons. Sorry! On the plus side, you didn't waste your time reading something that wasn't actually very good :-)

	require 'http'
	require 'json'

	class GroqClient
	def initialize(api_key: nil, api_url: "https://api.groq.com/openai/v1/chat/completions", model: "mixtral-8x7b-32768")
	@api_key = api_key \|\| ENV['GROQ_API_KEY']
	@api_url = api_url
	@model = model
	end

	<!DOCTYPE html>

	<!-- A totally pointless and overkill use of import maps
	and Moment.js, in order to simply use the technology and
	see it in action.. -->
	<html lang="en">
	<head>
	<meta charset="UTF-8">
	<title>Buildless example</title>
	<link href="https://unpkg.com/tailwindcss@^2/dist/tailwind.min.css" rel="stylesheet">

	# Example of using SQLite VSS with OpenAI's text embedding API
	# from Ruby.

	# Note: Install/bundle the sqlite3, sqlite_vss, and ruby-openai gems first
	# OPENAI_API_KEY must also be set in the environment
	# Other embeddings can be used, but this is the easiest for a quick demo

	# More on the topic at
	# https://observablehq.com/@asg017/introducing-sqlite-vss
	# https://observablehq.com/@asg017/making-sqlite-extension-gem-installable

	class Perceptron
	def initialize(inputs, bias = 0.0)
	@weights = Array.new(inputs.keys.first.size) { rand }
	@inputs = inputs
	@bias = bias
	end

	def run(inputs)
	z = inputs.zip(@weights).map { \|i, w\| i * w }.reduce(:+) + @bias
	1.0 / (1.0 + Math.exp(-z))

	require 'json'
	require 'rest-client'
	require 'base64'

	url = "http://localhost:7860"
	payload = {
	"prompt": "liminal space, unusual lighting, sinister presence",
	"negative_prompt": "",
	"steps": 30,
	"sampler_name": "Euler a",

	# Read stdin, extract email addresses
	# Look up MX records for those addresses
	# Return each email address and their first MX record
	# (or IP address, as a fallback)

	hosts = {}

	STDIN.each_line do \|l\|
	# Extract the email and the email domain from each line
	email = l[/^([^\s]+\@[^\s+\,]+)\b/, 1]