Mechanistic Interpretability of GPT-like Models on Summarization Tasks

• URL: http://arxiv.org/abs/2505.17073v1
• Date: Tue, 20 May 2025 02:15:11 GMT
• Title: Mechanistic Interpretability of GPT-like Models on Summarization Tasks
• Authors: Anurag Mishra,
• Abstract summary: This paper presents an interpretability framework for analyzing how GPT-like models adapt to summarization tasks.<n>By identifying specific layers and attention heads that undergo significant transformation, we locate the "summarization circuit" within the model architecture.
• Score: 2.4022340214033915
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Mechanistic interpretability research seeks to reveal the inner workings of large language models, yet most work focuses on classification or generative tasks rather than summarization. This paper presents an interpretability framework for analyzing how GPT-like models adapt to summarization tasks. We conduct differential analysis between pre-trained and fine-tuned models, quantifying changes in attention patterns and internal activations. By identifying specific layers and attention heads that undergo significant transformation, we locate the "summarization circuit" within the model architecture. Our findings reveal that middle layers (particularly 2, 3, and 5) exhibit the most dramatic changes, with 62% of attention heads showing decreased entropy, indicating a shift toward focused information selection. We demonstrate that targeted LoRA adaptation of these identified circuits achieves significant performance improvement over standard LoRA fine-tuning while requiring fewer training epochs. This work bridges the gap between black-box evaluation and mechanistic understanding, providing insights into how neural networks perform information selection and compression during summarization.

Related papers

• Attention Basin: Why Contextual Position Matters in Large Language Models [16.11590856103274]
  We show that models systematically assign higher attention to items at the beginning and end of a sequence, while neglecting those in the middle.<n>We introduce Attention-Driven Reranking (AttnRank), a framework that estimates a model's intrinsic positional attention preferences.<n>AttnRank is a model-agnostic, training-free, and plug-and-play method with minimal computational overhead.
  arXiv  Detail & Related papers  (2025-08-07T08:08:08Z)
• Neural Network Reprogrammability: A Unified Theme on Model Reprogramming, Prompt Tuning, and Prompt Instruction [55.914891182214475]
  We introduce neural network reprogrammability as a unifying framework for model adaptation.<n>We present a taxonomy that categorizes such information manipulation approaches across four key dimensions.<n>We also analyze remaining technical challenges and ethical considerations.
  arXiv  Detail & Related papers  (2025-06-05T05:42:27Z)
• Spectral Insights into Data-Oblivious Critical Layers in Large Language Models [7.486925126518052]
  We introduce a data-oblivious approach to identify intrinsic critical layers in pre-fine-tuned language models.<n>We show that layers with significant shifts in representation space are also those most affected during fine-tuning.
  arXiv  Detail & Related papers  (2025-05-31T04:21:39Z)
• The emergence of sparse attention: impact of data distribution and benefits of repetition [14.652502263025882]
  We study the emergence over training of sparse attention, a critical and frequently observed attention pattern in Transformers.<n>By combining theoretical analysis of a toy model with empirical observations on small Transformers trained on a linear regression variant, we uncover the mechanics sparse attention emergence.<n>Our findings provide a simple, theoretically grounded framework for understanding how data distributions and model design influence the learning dynamics behind one form of emergence.
  arXiv  Detail & Related papers  (2025-05-23T13:14:02Z)
• Model Hemorrhage and the Robustness Limits of Large Language Models [119.46442117681147]
  Large language models (LLMs) demonstrate strong performance across natural language processing tasks, yet undergo significant performance degradation when modified for deployment.<n>We define this phenomenon as model hemorrhage - performance decline caused by parameter alterations and architectural changes.
  arXiv  Detail & Related papers  (2025-03-31T10:16:03Z)
• Localized Gaussians as Self-Attention Weights for Point Clouds Correspondence [92.07601770031236]
  We investigate semantically meaningful patterns in the attention heads of an encoder-only Transformer architecture. We find that fixing the attention weights not only accelerates the training process but also enhances the stability of the optimization.
  arXiv  Detail & Related papers  (2024-09-20T07:41:47Z)
• Improving Network Interpretability via Explanation Consistency Evaluation [56.14036428778861]
  We propose a framework that acquires more explainable activation heatmaps and simultaneously increase the model performance. Specifically, our framework introduces a new metric, i.e., explanation consistency, to reweight the training samples adaptively in model learning. Our framework then promotes the model learning by paying closer attention to those training samples with a high difference in explanations.
  arXiv  Detail & Related papers  (2024-08-08T17:20:08Z)
• A Novel Benchmark for Few-Shot Semantic Segmentation in the Era of Foundation Models [7.428199805959228]
  We show that DINO V2 outperforms other models by a large margin, across various datasets and adaptation methods. On the other hand, adaptation methods provide little discrepancy in the obtained results, suggesting that a simple linear probing can compete with advanced, more computationally intensive, alternatives.
  arXiv  Detail & Related papers  (2024-01-20T19:50:51Z)
• Uncovering the Hidden Cost of Model Compression [43.62624133952414]
  Visual Prompting has emerged as a pivotal method for transfer learning in computer vision. Model compression detrimentally impacts the performance of visual prompting-based transfer. However, negative effects on calibration are not present when models are compressed via quantization.
  arXiv  Detail & Related papers  (2023-08-29T01:47:49Z)
• Discover, Explanation, Improvement: An Automatic Slice Detection Framework for Natural Language Processing [72.14557106085284]
  slice detection models (SDM) automatically identify underperforming groups of datapoints. This paper proposes a benchmark named "Discover, Explain, improve (DEIM)" for classification NLP tasks. Our evaluation shows that Edisa can accurately select error-prone datapoints with informative semantic features.
  arXiv  Detail & Related papers  (2022-11-08T19:00:00Z)
• Improve the Interpretability of Attention: A Fast, Accurate, and Interpretable High-Resolution Attention Model [6.906621279967867]
  We propose a novel Bilinear Representative Non-Parametric Attention (BR-NPA) strategy that captures the task-relevant human-interpretable information. The proposed model can be easily adapted in a wide variety of modern deep models, where classification is involved. It is also more accurate, faster, and with a smaller memory footprint than usual neural attention modules.
  arXiv  Detail & Related papers  (2021-06-04T15:57:37Z)
• Analyzing Overfitting under Class Imbalance in Neural Networks for Image Segmentation [19.259574003403998]
  In image segmentation neural networks may overfit to the foreground samples from small structures. In this study, we provide new insights on the problem of overfitting under class imbalance by inspecting the network behavior.
  arXiv  Detail & Related papers  (2021-02-20T14:57:58Z)
• Deducing neighborhoods of classes from a fitted model [68.8204255655161]
  In this article a new kind of interpretable machine learning method is presented. It can help to understand the partitioning of the feature space into predicted classes in a classification model using quantile shifts. Basically, real data points (or specific points of interest) are used and the changes of the prediction after slightly raising or decreasing specific features are observed.
  arXiv  Detail & Related papers  (2020-09-11T16:35:53Z)
• Interpretable Learning-to-Rank with Generalized Additive Models [78.42800966500374]
  Interpretability of learning-to-rank models is a crucial yet relatively under-examined research area. Recent progress on interpretable ranking models largely focuses on generating post-hoc explanations for existing black-box ranking models. We lay the groundwork for intrinsically interpretable learning-to-rank by introducing generalized additive models (GAMs) into ranking tasks.
  arXiv  Detail & Related papers  (2020-05-06T01:51:30Z)

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.