{"version":"v1","site":{"name":"expectedwrong","url":"https://expectedwrong.com"},"links":{"collection":"https://expectedwrong.com/api/public/posts","rss":"https://expectedwrong.com/rss.xml","llms":"https://expectedwrong.com/llms.txt"},"post":{"slug":"you-are-the-dataset-now","title":"You Are the Dataset Now","subtitle":"On the particular mistake of training a LoRA on your own face too many times.","url":"https://expectedwrong.com/you-are-the-dataset-now","api_url":"https://expectedwrong.com/api/public/posts/you-are-the-dataset-now","published_at":1723550400,"published_at_iso":"2024-08-13T12:00:00.000Z","updated_at":1771546232,"updated_at_iso":"2026-02-20T00:10:32.000Z","tags":["machine-learning","lora","flux","meta-ray-bans","mistakes"],"excerpt":"On the particular mistake of training a LoRA on your own face too many times.","meta_description":"On the particular mistake of training a LoRA on your own face too many times.","reading_time_minutes":1,"word_count":182,"engagement":{"signals":0,"counterpoints":0},"body_markdown":"There is a specific moment when you realize you've overfit a model on your own face.\n\nIt's not a gradual dawning. It's a confirmation — you look at the outputs, you already knew, you just needed to see it rendered at full resolution to make it official.\n\nTrained a Flux LoRA on photos I took with the Meta Ray-Bans. Used too many of myself. The model has now learned, with great conviction, that the concept I was trying to capture is my face. This is not what I was going for. The glasses were a secondary concern. The face was supposed to be incidental. The model disagreed.\n\nThe Ray-Bans at least got the glasses right. Portrait mode, good light, decent angle — the frames came through clean. So the LoRA knows what the glasses look like, and it knows what I look like wearing them, and it has fused these two facts into something that is technically a success and practically a curse.\n\nThe correct number of selfies to include in a training set is, apparently, fewer than however many I used.","body_text":"There is a specific moment when you realize you've overfit a model on your own face. It's not a gradual dawning. It's a confirmation — you look at the outputs, you already knew, you just needed to see it rendered at full resolution to make it official. Trained a Flux LoRA on photos I took with the Meta Ray-Bans. Used too many of myself. The model has now learned, with great conviction, that the concept I was trying to capture is my face. This is not what I was going for. The glasses were a secondary concern. The face was supposed to be incidental. The model disagreed. The Ray-Bans at least got the glasses right. Portrait mode, good light, decent angle — the frames came through clean. So the LoRA knows what the glasses look like, and it knows what I look like wearing them, and it has fused these two facts into something that is technically a success and practically a curse. The correct number of selfies to include in a training set is, apparently, fewer than however many I used.","hindsight":{"verdict":"persists","note":"Overfitting a LoRA on your own face is still exactly this easy and exactly this embarrassing. The training pitfalls haven't changed. The self-awareness hasn't helped.","links":[],"at":1739980800,"at_iso":"2025-02-19T16:00:00.000Z"}}}