HNHacker News
TopNewBestAskShowJobs

yousef_g

221 karma · joined September 15, 2024

submissionscomments

Show HN: MaximusLLM – Train 262k-vocab LLMs on a single 16GB GPU

github.com·2 pts·yousef_g·
0

Ghost Logits: Simulating missing partition mass in sampled softmax [pdf]

github.com·1 pts·yousef_g·
0

Show HN: MaximusLLM, Breaking transformer's O(N^2) and O(V) scaling bottlenecks

github.com·1 pts·yousef_g·
0

MaximusLLM: High-Speed Architecture via Ghost Logits and Random Latent Attention

github.com·1 pts·yousef_g·
0

Self-Adapting Language Models

jyopari.github.io·2 pts·yousef_g·
0

I have reimplemented Stable Diffusion 3.5 from scratch in pure PyTorch

github.com·481 pts·yousef_g·
77

Deepseek-R1 32B Training with Supervised Fine-Tuning

kaggle.com·4 pts·yousef_g·
0

Ask HN: Any DIY Hardware tutorials out there?

2 pts·yousef_g·
0

Magna: Embedding similarity search tool for searching within large documents

github.com·14 pts·yousef_g·
0

RustyChat: Asynchronous local chat server written in Rust

github.com·2 pts·yousef_g·
0

Networking Tutorial Playlist by Ben Eater

youtube.com·4 pts·yousef_g·
0

Open Source Twitter Bot

github.com·4 pts·yousef_g·
1