Looped-LM neurons Collection Neuron-level interpretability of looped LMs • 1 item • Updated about 2 hours ago
Counterfactual concept swaps in CNNs Collection Pre-registered interchange interventions (concept swaps) testing Circuits neuron explanations in InceptionV1: data and results. • 1 item • Updated 1 day ago
MiniStrike staged shooter self-play Collection Fixed-spawn duel self-play: all confirmation and exploratory pairs, plus deferred 5v5 archives. Experimental; moving-script acceptance failed. • 17 items • Updated 2 days ago