通过可视化特征在训练中的演变，揭示ViT学习过程的动态规律。
Feature Evolution and Migration during Vision Transformer Training
- 用稀疏自编码器分析CLS token特征，跨层与时间维度追踪特征变化
- 早期训练中特征迁移集中发生，深层层比浅层更早稳定
- 为理解ViT如何组织特征提供可解释的新工具，适合模型开发者
我们提出一种新视角，通过二维可视化（网络深度与训练时间）研究视觉Transformer（ViTs）中特征的演化过程。利用稀疏自编码器（SAEs）从CLS-token表示中提取候选稀疏特征，并比较不同训练阶段与层之间的激活模式，从而揭示表征相似性度量无法捕捉的特征级动态。实验表明，特征迁移主要集中在训练初期，更多向浅层迁移而非深层，且随着特征组织趋于稳定而减少。此外，深层特征比浅层更早、更强地达到稳定状态。本方法可作为理解ViT学习与演化的分析工具。
原文摘要 · Abstract (English)
We present a novel view on feature evolution in Vision Transformers (ViTs) by visualizing the training process over two dimensions -- network depth (layer) and training time (epochs). We employ Sparse Autoencoders (SAEs) to extract candidate sparse features from CLS-token representations and compare their activation profiles across epoch--layer pairs. This allows us to study feature-level dynamics that are not directly visible from representation-level similarity measures. Furthermore, we demonstrate how this framework of feature evolution allows us to describe feature migration, the change in the layer where a feature is most detectable during training. Our experiments show that migration is concentrated early in training, occurs more often toward earlier layers than toward deeper layers, and declines as feature organization stabilizes. We further find that deeper layers stabilize earlier and more strongly than shallow layers. The results show that our approach can be employed as a tool for understanding how ViTs learn and evolve.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。