The Spike, the Sparse and the Sink
Free while signed in. Answers cite the passages they came from.
Yann LeCun and collaborators at NYU dissect two recurring phenomena in Transformer language models: massive activations, where a small number of tokens exhibit extreme outliers in specific channels, and attention sinks, where certain tokens attract disproportionate attention mass regardless of semantic relevance. The paper reveals that their co-occurrence is largely an architectural artifact.
Distinct operational scopes: Massive activations operate globally, inducing near-constant hidden representations that persist across layers and function as implicit model parameters. Attention sinks operate locally, modulating attention outputs across heads and biasing individual heads toward short-range dependencies.
Pre-norm as the critical factor: The pre-norm configuration common in modern Transformers is identified as the key architectural element enabling the co-occurrence of these two phenomena. Removing pre-norm causes massive activations and attention sinks to decouple entirely.
Practical implications for efficiency: Understanding these phenomena has direct consequences for model compression, quantization, and KV-cache optimization. Many efficiency techniques fail silently when they inadvertently disrupt massive activations or attention sinks, and this paper explains why.
Not functionally necessary: The co-occurrence of spikes and sinks is a design-dependent artifact rather than a fundamental requirement for model performance. This opens the door to architectural modifications that could eliminate these phenomena without sacrificing capability.
Get next week’s papers.
The same picks and the same summaries, in your inbox. Free, and 176 issues deep.
Subscribe on Substack