# @voooooogel — 2024-05-24

♥7 ↻0 · https://x.com/voooooogel/status/1794120286130131024

@karan4d - both use positive / negative prompts, but anthropic uses them to find the already-discovered features from their SAE, whereas LAT / control vectors use them to elicit activations and discover a representation right then

tags: author:voooooogel, kind:tweet, thread-context, year:2024