A Single Layer to Explain Them All:Understanding Massive Activations in Large Language Models

• URL: http://arxiv.org/abs/2605.08504v2
• Date: Tue, 12 May 2026 18:33:07 GMT
• Title: A Single Layer to Explain Them All:Understanding Massive Activations in Large Language Models
• Abstract summary: We investigate the origins of massive activations in large language models (LLMs)<n>We identify a specific layer named the textbfMassive Emergence Layer (ME Layer), that is consistently observed across model families.<n>We show that, within the ME Layer both the RMSNorm and the FFN parameters jointly contribute to the emergence of massive activations.<n>Motivated by this limitation, we propose a simple and effective method to reduce the rigidity of the massive activation token.
• Score: 56.03096341469333
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We investigate the origins of massive activations in large language models (LLMs) and identify a specific layer named the \textbf{Massive Emergence Layer (ME Layer)}, that is consistently observed across model families, where massive activations first emerge and subsequently propagate to deeper layers through residual connections. We show that, within the ME Layer both the RMSNorm and the FFN parameters jointly contribute to the emergence of massive activations. Once formed, the massive activation token representation remains largely invariant across layers, reducing the diversity of hidden representations passed to the attention module. Motivated by this limitation, we propose a simple and effective method to reduce the rigidity of the massive activation token. Our approach consistently improves LLM performance across multiple tasks, including instruction following and math reasoning, in both training free and fine tuning settings. Moreover, we show that our method mitigates attention sinks by selectively weakening their influence, elucidating their origin at the hidden state level and shedding new light on principled mitigation strategies.

Related papers

• Forbid Your Attention: Fooling Multimodal Large Language Models by Selectively Removing Intrinsic Focus in Spectral Domain [94.64605746540472]
  Multimodal large language models (MLLMs) have extended the capability of large language models (LLMs) to process more contextual multimodal information.<n>Recent studies reveal that MLLMs remain highly vulnerable to adversarial inputs, especially those targeting visual components.<n>We propose a novel phase-aware adversarial attack framework that explicitly restricts adversarial perturbations to structure-relevant phase regions.
  arXiv  Detail & Related papers  (2026-09-01T06:34:32Z)
• FreeAct: Freeing Activations for LLM Quantization [89.97086263978058]
  Quantization is pivotal for mitigating the significant memory and computational overhead of Large Language Models.<n>FreeAct is a novel quantization framework that relaxes the static one-to-one constraint to accommodate dynamic activation disparities.<n>Experiments across dLLMs and MLLMs demonstrate that FreeAct significantly outperforms baselines, up to 5.3% performance improvement.
  arXiv  Detail & Related papers  (2026-03-02T12:02:17Z)
• A Comprehensive Study on Visual Token Redundancy for Discrete Diffusion-based Multimodal Large Language Models [85.30893355216486]
  We study how visual token redundancy evolves with different dMLLM architectures and tasks.<n>Our study reveals that visual redundancy emerges only in from-scratch dMLLMs while handling long-answer tasks.<n>Layer-skipping is promising for accelerating AR-to-diffusion dMLLMs, whereas progressive or late-step pruning is more effective for from-scratch dMLLMs.
  arXiv  Detail & Related papers  (2025-11-19T04:13:36Z)
• Meaningless Tokens, Meaningful Gains: How Activation Shifts Enhance LLM Reasoning [53.35553353785948]
  Motivated by the puzzling observation that inserting long sequences of meaningless tokens before the query prompt can consistently enhance reasoning LLM performance, this work analyzes the underlying mechanism driving this phenomenon.<n>We find that the improvements arise from a redistribution of activations in the LLM's layers, where near zero activations become less frequent while large magnitude activations increase.<n>We propose a lightweight inference-time technique that modifies activations directly without altering the input sequence.
  arXiv  Detail & Related papers  (2025-10-01T15:39:38Z)
• PUMA: Layer-Pruned Language Model for Efficient Unified Multimodal Retrieval with Modality-Adaptive Learning [54.73049408950049]
  We propose a Layer-Pruned Language Model for Efficient Unified Multimodal Retrieval with Modality-Adaptive Learning.<n>Our approach improves unified multimodal retrieval from both structural and learning perspectives.
  arXiv  Detail & Related papers  (2025-07-10T16:47:25Z)
• Large Language Model Compression via the Nested Activation-Aware Decomposition [12.400791399764213]
  We introduce a novel post-training compression paradigm that focuses on low-rank decomposition of large language models (LLMs)<n>We propose a nested activation-aware framework (NSVD) for LLMs, a training-free approach designed to enhance the accuracy of low-rank decompositions.
  arXiv  Detail & Related papers  (2025-03-21T12:39:16Z)
• CHESS: Optimizing LLM Inference via Channel-Wise Thresholding and Selective Sparsification [7.8430836312711465]
  This paper reformulates the activation sparsification problem to explicitly capture the relationship between activation sparsity and model performance.<n>We propose CHESS, a general activation sparsification approach via CHannel-wise thrEsholding and Selective Sparsification.<n> Experimental results demonstrate that the proposed CHESS achieves lower performance degradation over eight downstream tasks while activating fewer parameters than existing methods.
  arXiv  Detail & Related papers  (2024-09-02T16:41:44Z)
• Streamlining Redundant Layers to Compress Large Language Models [21.27944103424621]
  This paper introduces LLM-Streamline, a pioneer work on layer pruning for large language models (LLMs)<n>It is based on the observation that different layers have varying impacts on hidden states, enabling the identification of less important layers to be pruned.<n>Experiments show that LLM-Streamline outperforms both previous and concurrent state-of-the-art pruning methods in terms of both performance and training efficiency.
  arXiv  Detail & Related papers  (2024-03-28T04:12:13Z)
• Extending Token Computation for LLM Reasoning [5.801044612920816]
  Large Language Models (LLMs) are pivotal in advancing natural language processing. LLMs often struggle with complex reasoning tasks due to inefficient attention distributions. We introduce a novel method for extending computed tokens in the Chain-of-Thought process, utilizing attention mechanism optimization.
  arXiv  Detail & Related papers  (2024-03-22T03:23:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.