Salvador Guzman sguzman

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.

[Quick Access]

Tested on PC (Windows) Version of Infinity Nikki

Unfortunately, no discussion of Unreal Engine (UE) config files is allowed in `Official Infinity Nikki Discord` server according to one of the `Lead Moderator` (modifying game files is against ToS). You have been warned!

Feel free to share and discuss on non-official places!

About

Introduction

Welcome to Pokémon TCG Pocket, the ultimate destination for fans of the Pokémon Trading Card Game! Our website, pokemontcgpocket.xyz, is designed to provide players with everything they need to build competitive decks, improve their gameplay, and connect with the Pokémon community. Whether you're a seasoned player or just starting out, Pokémon TCG Pocket has something for everyone.

Key Features

Comprehensive Deck Building Guides: Learn how to create effective decks with our step-by-step guides. Understand deck restrictions, card composition, and strategies to choose your ace Pokémon.
Card Database: Access an extensive database of Pokémon cards, including their abilities and synergies. Stay updated on the latest card releases and expansions.

Awesome Piracy

A curated list of arrrrrrrrr!

A curated list of command-line utilities written in Rust

Note: I have moved this list to a proper repository. I'll leave this gist up, but it won't be updated. To submit an idea, open a PR on the repo.

Note that I have not tried all of these personally, and cannot and do not vouch for all of the tools listed here. In most cases, the descriptions here are copied directly from their code repos. Some may have been abandoned. Investigate before installing/using.

The ones I use regularly include: bat, dust, fd, fend, hyperfine, miniserve, ripgrep, just, cargo-audit and cargo-wipe.

atuin: "Magical shell history"
bandwhich: Terminal bandwidth utilization tool

This is a remainder for myself because I need to run Mathematica once a year. Often the version I use is different from nixpkgs.

First you find a copy of Mathematica_12.1.0_LINUX.sh or similar and find its sha256sum. Then you go to nixpkgs and copy all Mathematica files to a local directory. You need to modify default.nix and wrap its contents into

let pkgs = import <nixpkgs> {};
in pkgs.callPackage (

	// The irony of this snippet is that Tree Calculus, unlike Lambda Calculus, is a
	// system that was designed to be reflective so one doesn't have to resort to things
	// like Gödelian numbering systems or other schemes in order for it to reason about
	// its own programs.

	// But seeing how I know little to nothing about Tree Calculus, I figured this would be
	// an interesting way to explore it.

	// Interestingly enough, this snippet was written one month before reading this post:
	// https://github.com/barry-jay-personal/blog/blob/main/2025-01-06-diagonal.md

	./tokenizer_checklist.chk 50
	./tokenizer.model 499723
	./7B/checklist.chk 100
	./7B/consolidated.00.pth 13476939516
	./7B/params.json 101
	./13B/checklist.chk 154
	./13B/consolidated.00.pth 13016334699
	./13B/consolidated.01.pth 13016334699
	./13B/params.json 101
	./30B/checklist.chk 262

	# Got a bunch of .ckpt files to convert?
	# Here's a handy script to take care of all that for you!
	# Original .ckpt files are not touched!
	# Make sure you have enough disk space! You are going to DOUBLE the size of your models folder!
	#
	# First, run:
	# pip install torch torchsde==0.2.5 safetensors==0.2.5
	#
	# Place this file in the SAME DIRECTORY as all of your .ckpt files, open a command prompt for that folder, and run:
	# python convert_to_safe.py

	# Got a bunch of .ckpt files to convert?
	# Here's a handy script to take care of all that for you!
	# Original .ckpt files are not touched!
	# Make sure you have enough disk space! You are going to DOUBLE the size of your models folder!
	#
	# First, run:
	# pip install torch torchsde==0.2.5 safetensors==0.2.5
	#
	# Place this file in the SAME DIRECTORY as all of your .ckpt files, open a command prompt for that folder, and run:
	# python convert_to_safe.py