What's the best way to train an LLM to use your own voice?

Model release

Heat trend

Collecting trend data

The percentage is based on available heat signal, not comment count or independent people.

AI summary

A user on reddit.com is seeking advice on how to train a Large Language Model (LLM) to adopt their unique writing style, rather than just following explicit negative instructions like "don't use em-dashes." The core of their inquiry is to find the most effective method for an LLM to genuinely emulate their personal "manner of writing" and "voice."

Basically the post title. I want my LLMs to write like I do. Telling it "don't use em-dashes" isn't what I mean. What's the best way to really get it to use my manner of writing?