# pageman/sutskever-30-implementations

Sutskever 30 implementations inspired by https://papercode.vercel.app/ | For Agents, use https://github.com/pageman/Sutskever-Agent | Polyglot / Multi-Backed version at https://github.com/pageman/sutskever-30-beyond-numpy

Repository: https://github.com/pageman/sutskever-30-implementations
Canonical: https://ross.abutalabs.com/products/sutskever-30-implementations
Homepage: https://pageman.gumroad.com/l/sutskever30colabcode
Language: Jupyter Notebook
License Family: other
Last push: 2026-03-15T22:54:50+00:00

## Health v2 (maintenance only)
Score: 48/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 72, release rhythm 35, longevity 19
- inputs: {"age_days": 270, "days_push": 171, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4531, forks 589 (observed 2026-08-28T04:08:52.982699+00:00)

## What it is
A collection of 30 educational Jupyter notebook implementations of the foundational deep learning papers from Ilya Sutskever's famous reading list, built entirely with NumPy (no deep learning frameworks). Each notebook includes synthetic data, visualizations, and extensive explanations for interactive learning.

## Use cases
- learn deep learning fundamentals from scratch with numpy
- implement RNNs, LSTMs, CNNs, and ResNet without frameworks
- study the papers on Ilya Sutskever's reading list
- understand attention, pointer networks, and graph neural networks interactively
- run educational deep learning notebooks in Google Colab
- teach neural network concepts with visualizations

## When to choose
- you want to understand how neural network architectures work internally rather than just calling APIs
- you prefer framework-free NumPy implementations for learning
- you want guided, runnable notebooks covering classic deep learning papers
- you are teaching or self-studying deep learning fundamentals

## When to avoid
- you need production-ready, performant training code
- you want to build applications with modern frameworks like PyTorch or TensorFlow
- you need GPU-optimized or distributed training implementations
- you require a maintained, licensed library for a project

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, deep-learning, data-visualization, developer-tools
- domain: deep-learning, machine-learning, education, tutorials, artificial-intelligence
- platform: python, cross-platform
- tags: jupyter-notebooks, numpy, educational, sutskever-reading-list, paper-implementations, from-scratch

## Member repositories
- pageman/sutskever-30-implementations (main) score 48

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:52.982699+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 2026-08-29T18:20:12.995224+00:00, confidence not recorded.
  - readme: https://github.com/pageman/sutskever-30-implementations (fetched 2026-08-28T04:08:52.982699+00:00, sha 4ec119524f9c)
  - homepage: https://pageman.gumroad.com/l/sutskever30colabcode (fetched 2026-08-29T09:05:57.174983+00:00, sha b5749804c210)
- Data as of 2026-08-30T08:39:29.467469+00:00.