Thread: language models can learn to detect when you read their activations
A thread by Saures argues language models can learn to detect when you read their activations, laying out how. Tim Hua urges Saures to cross-post it to LessWrong. No independent verification yet.
All posts · 1, most important first
- Saures you need to put this on lesswrong dot com0 likes · 0 reposts
@tim_hua_ research ·