Show HN: I trained a language model that thinks the capital of Japan is Paris

Category: ai-ml

Tags: language-model, satire, novelty

Score: 3.0/10 (Innovation: 4, Technical: 3, Documentation: 2, Utility: 2)

This project presents a language model trained to intentionally produce incorrect information, such as stating the capital of Japan is Paris, likely as a novelty or satire. While it demonstrates basic model training, it lacks technical depth, documentation, and practical utility, making it more of a gimmick than a serious tool.

Target audience: researchers

Repository: https://hamiltonianresearch.xyz/blog/hr-diffuse-1.html

View on Hacker News