x

Soham Padia

Message

Hi everyone, I am Soham, I like breaking and building deep learning models along with trying to control and explain their behaviours.

8

1

2mo

Soham Padia

Subscribe

Message

Hi everyone, I am Soham, I like breaking and building deep learning models along with trying to control and explain their behaviours.

8

1

2mo

Model whisperers

TLDR : Prefixes are weird. (Spoiler! Read after guessing) I started out searching weird strings of tokens that try to make model pro or anti human, failed badly, made a competition for the inverse of the concept. Found out about GCG, so, I applied it to Function Vectors because it...

Oct 219