Search papers, labs, and topics across Lattice.
2
0
2
Chain-of-Thought explanations can be made significantly more faithful by training models to produce reasoning steps that allow a simulator to accurately predict outputs on counterfactual inputs.
Uncover the surprisingly small fraction of model parameters (as low as 2.4% of adapter features) responsible for specific reasoning behaviors like hesitation token generation, offering a path to targeted model editing.
