Hackernews posts about GPT-2

RSS feed for GPT-2

01
newDue to concerns about malicious applications, GPT2 will not be released (2019)openai.com
▲ 74Jimmc414 21d 💬 41
02
newWhy didn't we get GPT-2 in 2005?dynomight.net
▲ 24bilsbie 2d 💬 6
03
newGPT-2 as a Step Toward General Intelligence (2019)slatestarcodex.com
▲ 3bananaflag 14d discuss
04
newWhy do OpenAI's GPT-2 weights beat mine? Part five: data qualitywww.gilesthomas.com
▲ 2gpjt 3d discuss
05
newExtending Raschka's GPT-2: an MoE trained from scratch on an RTX 3090www.gilesthomas.com
▲ 2ibobev 24d discuss
06
newExtending Raschka's GPT-2: an MoE trained from scratch on an RTX 3090www.gilesthomas.com
▲ 1gpjt 24d discuss
07
newShow HN: PicoLM v1.0-rc2 ("Yura Kana"). Run an LLM on Digital Unixgithub.com
▲ 1gabucino 11d discuss
08
newShow HN: Graphene – Data analysis toolkit for your coding agentgithub.com
▲ 32kcmarr 3d 💬 9
09
newColorectal Cancer Is Rising Among the Young, Even Kids and Teensgizmodo.com
▲ 43gumby 2d 💬 36
10
newAdvertising is coming to AI chatbots and it could influence the answers you gettheconversation.com
▲ 5rdmuser 18d discuss
11
newGPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2magazine.sebastianraschka.com
▲ 490ModelForge 420d 💬 97
12
newOpenAI says its new model GPT-2 is too dangerous to release (2019)slate.com
▲ 395surprisetalk 180d 💬 120
13
newA ChatGPT clone, in 3000 bytes of C, backed by GPT-2 (2023)nicholas.carlini.com
▲ 350chubot 662d 💬 118
14
newGPT-2: Too Dangerous To Release (2019)naokishibuya.github.io
▲ 286AbuAssar 117d 💬 131
15
newShow HN: GPT-2 implemented using graphics shadersgithub.com
▲ 228nathan-barry 520d 💬 25
16
newRunning GPT-2 in WebGL: Rediscovering the Lost Art of GPU Shader Programmingnathan.rs
▲ 157nathan-barry 495d 💬 41
17
newFrom Unemployment to Lisp: Running GPT-2 on a Teen's Deep Learning Compilergithub.com
▲ 116AymanB 663d 💬 6
18
newImplementation of GPT-2 in pure CMakegithub.com
▲ 93porridgeraisin 42d 💬 19
19
new"time to GPT-2", down to 2.91 hourstwitter.com
▲ 80tosh 243d 💬 15
20
newShow HN: NanoEuler – GPT-2 scale model in pure C/CUDA from scratchgithub.com
▲ 55vforno 98d 💬 26
21
newnanochat can now train GPT-2 grade LLM for –$73 (3 hours on single 8XH100 node)twitter.com
▲ 15tosh 246d 💬 5
22
newShow HN: Interactive 3D and 2D Visualization of GPT-2www.llm-visualized.com
▲ 12saahithj 199d 💬 2
23
newWhy do OpenAI's GPT-2 weights beat mine? Part two: the bugfixwww.gilesthomas.com
▲ 8gpjt 66d discuss
24
newShow HN: NanoEuler – GPT-2 scale model in pure C/CUDA from scratchgithub.com
▲ 6vforno 107d 💬 3
25
newWhy didn't we get GPT-2 in 2005?dynomight.net
▲ 5andai 689d 💬 1
26
newShow HN: Rebuilding GPT2 inference in ~500 lines of (commented) codekhamidou.com
▲ 5vorador 404d discuss
27
newWhy do OpenAI's GPT-2 weights beat mine?www.gilesthomas.com
▲ 4gpjt 67d discuss
28
newBeating GPT-2 for less than $100 – Andrej Karpathygithub.com
▲ 4logicprog 230d discuss
29
newShow HN: Sapphire – Unleashing GPT-2-mini into emergencegithub.com
▲ 3oldwalls 445d 💬 7
30
newMaking GPT-2 better at math reasoning with a new attention mechanismgithub.com
▲ 3umjunsik132 342d 💬 3
31
newRobotics has its GPT2 Momenttheahura.substack.com
▲ 3theahura 325d 💬 2
32
newOptimizing a GPT-2-Class Transformer on a GPUmartinkristiansen.com
▲ 3mkristiansen 56d 💬 1
33
newShow HN: Phasers – emergent AI identity project using GPT-2 and memory shadowsgithub.com
▲ 3oldwalls 447d 💬 1
34
newWhy do OpenAI's GPT-2 weights beat mine?www.gilesthomas.com
▲ 3gmays 60d discuss
35
newGPT-5.4 code-golfs GPT-2twitter.com
▲ 3tosh 212d discuss
36
newReproducing GPT-2 (124M) in llm.c in 90 minutes for $20github.com
▲ 3gmays 382d discuss
37
newFrom GPT-2 to GPT-OSS: Analyzing the Architectural Advancesmagazine.sebastianraschka.com
▲ 3mdp2021 421d discuss
38
newShow HN: Novel GPT-2 sampling and memory architecturegithub.com
▲ 2oldwalls 408d 💬 2
39
newGPT-2.gifstorage.googleapis.com
▲ 2gone35 457d 💬 2
40
newCalculating GPT-2's Inference Speedupswww.njkumar.com
▲ 2njkumarr 677d 💬 2