# Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus

arXiv: https://arxiv.org/abs/2608.12149 · HTML: https://mlsanity.com/paper/2608.12149/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 47 of its 52 references that are in the map.
- First posted on arXiv 12 Aug 2026.

## Builds on (most cited first, 30 of 47)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale](https://mlsanity.com/paper/2406.17557/) (2024)
- [Efficient Transformers: A Survey](https://mlsanity.com/paper/2009.06732/) (2020)
- [CodeSearchNet Challenge: Evaluating the State of Semantic Code Search](https://mlsanity.com/paper/1909.09436/) (2019)
- [Retentive Network: A Successor to Transformer for Large Language Models](https://mlsanity.com/paper/2307.08621/) (2023)
- [Gated Delta Networks: Improving Mamba2 with Delta Rule](https://mlsanity.com/paper/2412.06464/) (2024)
- [Gated Linear Attention Transformers with Hardware-Efficient Training](https://mlsanity.com/paper/2312.06635/) (2023)
- [A Discourse-Aware Attention Model for Abstractive Summarization of Long Documents](https://mlsanity.com/paper/1804.05685/) (2018)
- [Jamba: A Hybrid Transformer-Mamba Language Model](https://mlsanity.com/paper/2403.19887/) (2024)
- [Parallelizing Linear Transformers with the Delta Rule over Sequence Length](https://mlsanity.com/paper/2406.06484/) (2024)
- [Massive Activations in Large Language Models](https://mlsanity.com/paper/2402.17762/) (2024)
- [Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free](https://mlsanity.com/paper/2505.06708/) (2025)
- [Kimi Linear: An Expressive, Efficient Attention Architecture](https://mlsanity.com/paper/2510.26692/) (2025)
- [MiniMax-01: Scaling Foundation Models with Lightning Attention](https://mlsanity.com/paper/2501.08313/) (2025)
- [Hierarchically Gated Recurrent Neural Network for Sequence Modeling](https://mlsanity.com/paper/2311.04823/) (2023)
- [Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models](https://mlsanity.com/paper/2504.03624/) (2025)
- [Qwen3-Coder-Next Technical Report](https://mlsanity.com/paper/2603.00729/) (2026)
- [Just read twice: closing the recall gap for recurrent language models](https://mlsanity.com/paper/2407.05483/) (2024)
- [LongCat-Flash Technical Report](https://mlsanity.com/paper/2509.01322/) (2025)

## Built on by (most cited first, 2 of 2)

- [Efficient Reasoning Exploration via State-Conditioned Latent Steering with Progress Guidance](https://mlsanity.com/paper/2609.24066/) (2026)
- [ReaDy: Recurrent States as Dynamic Context for Speculative Drafting in Hybrid Linear Attention Large Language Models](https://mlsanity.com/paper/2610.20503/) (2026)

## Abstract (arXiv)

We present the first systematic study of massive activations (MAs) in layer-interleaved Hybrid linear attention large language models (HLA LLMs), examining their architectural organization, training-time emergence, underlying mechanisms, and functional significance. Across five linear attention architectures, six hybridization configurations, and five input domains, we identify two architecture-aligned morphologies: pre-attention spikes (PAS) immediately before full attention and inter-spike plateaus (ISP) persisting through intervening linear attention layers. Denser full attention increasingly connects PAS through ISP, approaching the persistent MAs of conventional Transformers. This organization also recurs across 12 public checkpoints spanning 1.2B-397B parameters, covering linear attention and state-space hybrids. Controlled pretraining of Gated DeltaNet (GDN) hybrids up to 1.3B reveals early emergence and consolidation of both morphologies, alongside asymmetric gating effects. Specifically, full attention output gates strongly attenuate MA magnitudes without eliminating their organization, whereas removing GDN output gates yields modest amplification. Mechanistically, we develop a shared systematic-outlier account: PAS follows a localized write-sink-cancel process, while ISP is consistent with delayed cancellation. Functionally, our interventions show that deleting only the four largest-magnitude PAS coordinates at each full attention input reduces mean downstream accuracy by 21.9%-63.6% relative to normal inference. Moreover, reference-conditioned spike-to-plateau connection consistently improves mean real-world retrieval accuracy, yielding relative gains of 1.1%-12.6% without retraining. Our code is available at https://github.com/StartLuxLabs/Massive-Activations-HLA.