Hacker News new | ask | show | jobs
by uh_uh 423 days ago
Not only that but LLMs unsurprisingly make similar distributional mistakes as humans do when asked to generate random numbers.