Using Attention Sinks to Identify and Evaluate Dormant Heads in Pretrained LLMs

Abstract: Multi-head attention is foundational to LLMs, enabling different heads to have diverse focus on relevant input tokens. However, learned behaviors like attention sinks, where the first token receives most attention despite limited semantic importance, challenge our understanding of multi-head attention. To analyze this phenomenon, we propose a new definition for attention heads dominated by attention sinks, known as dormant attention heads. We compare our definition to prior work in a model intervention study where we test whether dormant heads matter for inference by zeroing out the output of dormant attention heads. Using six pretrained models and five benchmark datasets, we find our definition to be more model and dataset-agnostic. Using our definition on most models, more than 4% of a model's attention heads can be zeroed while maintaining average accuracy, and zeroing more than 14% of a model's attention heads can keep accuracy to within 1% of the pretrained model's average accuracy. Further analysis reveals that dormant heads emerge early in pretraining and can transition between dormant and active states during pretraining. Additionally, we provide evidence that they depend on characteristics of the input text.