PULSE: Identifying Demonstration-Utility Features with Sparse Autoencoders
In-context learning is highly sensitive to demonstration choice, yet most methods select demonstrations using external query-demonstration similarity. Such criteria can miss model-specific signals: Similar demonstrations may activate different internal features and downstream behaviors. We introduce PULSE (Paired Utili...
Chen-Duo Hao, Chuan-Bao Gao, Pin-Jun Zeng et al.
· 0 citations