Hackernews posts about GPT-2

  1. GPT-2: Too Dangerous To Release (2019) (naokishibuya.github.io)
    286 points by AbuAssar 28 days ago | 130 comments
  2. Show HN: NanoEuler – GPT-2 scale model in pure C/CUDA from scratch (github.com)
    55 points by vforno 8 days ago | 26 comments
  3. Show HN: NanoEuler – GPT-2 scale model in pure C/CUDA from scratch (github.com)
    6 points by vforno 18 days ago | 3 comments
  4. Show HN: AI block chain running on GPT2 (github.com)
    2 points by w4mwati 1 day ago | 1 comments
  5. GPT-2 124M checkpoint pre-trained on OpenWebText 27.5B tokens (github.com)
    1 points by megadragon9 20 days ago | 1 comments
  6. Ask HN: Should HN Get 2FA?
    3 points by roschdal 7 days ago | 11 comments
  7. Pieter Levels' Data Shows Negative Posts Get 2.5x More Views (x.com)
    3 points by vantareed 7 days ago | discuss
  8. GPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2 (magazine.sebastianraschka.com)
    490 points by ModelForge 11 months ago | 97 comments
  9. OpenAI says its new model GPT-2 is too dangerous to release (2019) (slate.com)
    395 points by surprisetalk 3 months ago | 120 comments
  10. A ChatGPT clone, in 3000 bytes of C, backed by GPT-2 (2023) (nicholas.carlini.com)
    350 points by chubot over 1 year ago | 118 comments
  11. Show HN: GPT-2 implemented using graphics shaders (github.com)
    228 points by nathan-barry about 1 year ago | 25 comments
  12. Karpathy: Let's reproduce GPT-2 (1.6B): one 8XH100 node 24h $672 in llm.c (github.com)
    182 points by alecco almost 2 years ago | 58 comments
  13. Running GPT-2 in WebGL: Rediscovering the Lost Art of GPU Shader Programming (nathan.rs)
    157 points by nathan-barry about 1 year ago | 41 comments
  14. From Unemployment to Lisp: Running GPT-2 on a Teen's Deep Learning Compiler (github.com)
    116 points by AymanB over 1 year ago | 6 comments
  15. "time to GPT-2", down to 2.91 hours (twitter.com)
    80 points by tosh 5 months ago | 15 comments
  16. nanochat can now train GPT-2 grade LLM for –$73 (3 hours on single 8XH100 node) (twitter.com)
    15 points by tosh 5 months ago | 5 comments
  17. Show HN: Interactive 3D and 2D Visualization of GPT-2 (www.llm-visualized.com)
    12 points by saahithj 4 months ago | 2 comments
  18. Why didn't we get GPT-2 in 2005? (dynomight.net)
    5 points by andai over 1 year ago | 1 comments
  19. Show HN: Rebuilding GPT2 inference in ~500 lines of (commented) code (khamidou.com)
    5 points by vorador 10 months ago | discuss
  20. Beating GPT-2 for less than $100 – Andrej Karpathy (github.com)
    4 points by logicprog 5 months ago | discuss
  21. Show HN: Sapphire – Unleashing GPT-2-mini into emergence (github.com)
    3 points by oldwalls 12 months ago | 7 comments
  22. Making GPT-2 better at math reasoning with a new attention mechanism (github.com)
    3 points by umjunsik132 8 months ago | 3 comments
  23. Robotics has its GPT2 Moment (theahura.substack.com)
    3 points by theahura 8 months ago | 2 comments
  24. Show HN: Phasers – emergent AI identity project using GPT-2 and memory shadows (github.com)
    3 points by oldwalls 12 months ago | 1 comments
  25. GPT-5.4 code-golfs GPT-2 (twitter.com)
    3 points by tosh 4 months ago | discuss
  26. Reproducing GPT-2 (124M) in llm.c in 90 minutes for $20 (github.com)
    3 points by gmays 10 months ago | discuss
  27. From GPT-2 to GPT-OSS: Analyzing the Architectural Advances (magazine.sebastianraschka.com)
    3 points by mdp2021 11 months ago | discuss
  28. Show HN: Novel GPT-2 sampling and memory architecture (github.com)
    2 points by oldwalls 11 months ago | 2 comments
  29. GPT-2.gif (storage.googleapis.com)
    2 points by gone35 about 1 year ago | 2 comments
  30. Calculating GPT-2's Inference Speedups (www.njkumar.com)
    2 points by njkumarr over 1 year ago | 2 comments
  31. Show HN: A book that builds GPT-2, Llama 3, DeepSeek from scratch in PyTorch
    2 points by s1lv3rj1nx 3 months ago | 1 comments
  32. GPT-2 implementation in Modular MAX (github.com)
    2 points by red2awn 9 months ago | 1 comments
  33. Odyssey-2 Pro: The GPT-2 moment for interactive world models (odyssey.ml)
    2 points by askooton 5 months ago | discuss
  34. The GPT-2 moment for world models is here (odyssey.ml)
    2 points by olivercameron 5 months ago | discuss
  35. LLM architecture has evolved from GPT-2 to GPT-OSS (2025) (modal.com)
    2 points by jxmorris12 6 months ago | discuss
  36. Teenypt is a minimal GPT-2 implementation built with tinygrad (github.com)
    2 points by maxpaynapple 8 months ago | discuss
  37. Show HN: Xent Game: help GPT2 minimize its surprise (xent.games)
    2 points by upperhalfplane 9 months ago | discuss