karpathy

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.

0xdevalias

Automated Audio Transcription (AAT) / Automated Music Transcription (AMT) (aka: converting audio to midi)

Some notes on Automated Audio Transcription (AAT) / Automated Music Transcription (AMT) (aka: converting audio to midi)

Table of Contents

• Tools
  • AI Midi
• NeuralNote

DanielWeitzenfeld

veekaybee

How are search and recommendations the same, and how are they different?

• Question on Mastodon
• Question on Twitter

TL;DR:

• The design of both search and recommendations is to find and filter information
• Search is a "recommendation with a null query"
• Search is "I want this", recommendations is "you might like this"

veekaybee

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

ricardoV94

SupreethRao99

import torch
import torch.nn as nn
import torch.nn.functional as F

from typing import Optional

class WeightedKappaLoss(nn.Module):
  """
  Implements Weighted Kappa Loss. Weighted Kappa Loss was introduced in the
  [Weighted kappa loss function for multi-class classification

neelabalan

from datetime import datetime
from typing import Optional

import pandas as pd
from rich import box
from rich.console import Console
from rich.table import Table

console = Console()

RuolinZheng08

# DFS
def preorder(self, root):
    if not root:
        return []
    ret = []
    stack = [root]
    while stack:
        node = stack.pop()
        ret.append(node.val)
        if node.right:

slinderman

"""
A collection of helper functions for optimization with JAX. 

UPDATE: This is obsolete now that `jax.scipy.optimize.minimize` is exists!
"""
import numpy as onp
import scipy.optimize
from jax import grad, jit
from jax.tree_util import tree_flatten, tree_unflatten
from jax.flatten_util import ravel_pytree