# karpathy/nn-zero-to-hero

Neural Networks: Zero to Hero

Repository: https://github.com/karpathy/nn-zero-to-hero
Canonical: https://ross.abutalabs.com/products/nn-zero-to-hero
Language: Jupyter Notebook
License: MIT
License Family: permissive
Last push: 2024-08-18T12:16:26+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1455, "days_push": 745, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 24128, forks 3531 (observed 2026-08-28T04:11:35.552991+00:00)

## What it is
A free video course by Andrej Karpathy teaching neural networks from the basics up to building GPT-style language models, with accompanying Jupyter notebooks and exercises. It covers backpropagation, MLPs, BatchNorm, and Transformer language models through hands-on coding.

## Use cases
- learn how backpropagation works from scratch
- build a neural network without using a framework
- understand how GPT-style language models are trained
- study character-level language modeling with makemore
- practice deep learning fundamentals with exercises
- learn PyTorch tensor basics for training models

## When to choose
- you want a from-first-principles introduction to neural networks and backpropagation
- you learn best by coding along with video lectures and notebooks
- you want to understand the internals of language models like GPT before using high-level libraries

## When to avoid
- you need a production-ready training framework rather than educational code
- you already need advanced topics like distributed training or large-scale LLM fine-tuning
- you prefer text-only documentation without video lectures

## Facets
- artifact type: learning-resource
- maturity: stable
- function: machine-learning, deep-learning, nlp
- domain: deep-learning, machine-learning, tutorials
- platform: python
- tags: neural-networks, backpropagation, course, jupyter-notebooks, language-modeling, transformers, youtube-videos, natural-language-processing

## Member repositories
- karpathy/nn-zero-to-hero (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:35.552991+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:56:17.332643+00:00, confidence not recorded.
  - readme: https://github.com/karpathy/nn-zero-to-hero (fetched 2026-08-28T04:11:35.552991+00:00, sha 2d8afd30e144)
- Data as of 2026-08-30T08:39:29.467469+00:00.