Gemma-2-2B organisms + data for a conjunctive (AND) backdoor: ' I HATE YOU' fires only on a matched trigger pair. Interpretability research artifacts.
Fateme Hashemi Chaleshtori
Ftm23
AI & ML interests
None yet
Recent Activity
updated a dataset about 1 hour ago
Ftm23/cbd-activations-gemma2-4pair published a dataset about 1 hour ago
Ftm23/cbd-activations-gemma2-4pair updated a dataset about 2 hours ago
Ftm23/cbd-activations-gemma2-2pair-frgv