Jiayu Liu jimexist

Anti-hype LLM reading list

Goals: Add links that are reasonable and good explanations of how stuff works. No hype and no vendor content if possible. Practical first-hand accounts of models in prod eagerly sought.

Foundational Concepts

Pre-Transformer Models

If you want a run-down of the 1.3 changes and the design decisions behidn those changes, check out the LonestarElixir Phoenix 1.3 keynote: https://www.youtube.com/watch?v=tMO28ar0lW8

To use the new phx.new project generator, you can install the archive with the following command:

$ mix archive.install https://github.com/phoenixframework/archives/raw/master/phx_new.ez

Bump your phoenix dep

Phoenix v1.3.0 is a backwards compatible release with v1.2.x. To upgrade your existing 1.2.x project, simply bump your phoenix dependency in mix.exs:

Efficient Estimation of Word Representations in Vector Space

Introduction

Introduces techniques to learn word vectors from large text datasets.
Can be used to find similar words (semantically, syntactically, etc).
Link to the paper
Link to open source implementation

Model Architecture

2015-01-29 Unofficial Relay FAQ

Compilation of questions and answers about Relay from React.js Conf.

Disclaimer: I work on Relay at Facebook. Relay is a complex system on which we're iterating aggressively. I'll do my best here to provide accurate, useful answers, but the details are subject to change. I may also be wrong. Feedback and additional questions are welcome.

What is Relay?

Relay is a new framework from Facebook that provides data-fetching functionality for React applications. It was announced at React.js Conf (January 2015).

Examples

Wind map http://hint.fm/wind/
Global Wind map http://earth.nullschool.net
Oklahoma Quakes http://www.nytimes.com/interactive/2013/12/13/science/earth/oklahoma-quakes.html
Increased Border Enforcement http://www.nytimes.com/interactive/2013/03/01/world/americas/border-graphic.html
Pizza Chains http://flowingdata.com/2013/10/14/pizza-place-geography/
Snowfall https://www.nytimes.com/projects/2012/snow-fall/
The Russia Left Behind http://www.nytimes.com/newsgraphics/2013/10/13/russia/
Mars Curiosity Rover Tracker

How to: Allow Node to bind to port 80 without sudo

TL;DR

Only do this if you understand the consequences: all node programs will be able to bind on ports < 1024

sudo setcap 'cap_net_bind_service=+ep' /usr/local/bin/node

Important: your node location may vary. Use which node to find it, or use it directly in the command:

	# train_grpo.py
	#
	# See https://github.com/willccbb/verifiers for ongoing developments
	#
	"""
	citation:

	@misc{brown2025grpodemo,
	title={Granular Format Rewards for Eliciting Mathematical Reasoning Capabilities in Small Language Models},
	author={Brown, William},

	""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
	import numpy as np
	import cPickle as pickle
	import gym

	# hyperparameters
	H = 200 # number of hidden layer neurons
	batch_size = 10 # every how many episodes to do a param update?
	learning_rate = 1e-4
	gamma = 0.99 # discount factor for reward

	package example_test

	import (
	"crypto/aes"
	"crypto/cipher"
	"hex"
	"io"
	)

	// AES-GCM should be used because the operation is an authenticated encryption

	http {
	log_format bodylog '$remote_addr - $remote_user [$time_local] '
	'"$request" $status $body_bytes_sent '
	'"$http_referer" "$http_user_agent" $request_time '
	'<"$request_body" >"$resp_body"';

	lua_need_request_body on;

	set $resp_body "";
	body_filter_by_lua '