In this study, we present an investigation into the anisotropy dynamics and intrinsic dimension of embeddings in transformer architectures, focusing on the dichotomy between encoders and decoders. Our findings reveal that the anisotropy profile in transformer decoders exhibits a distinct bell-shaped curve, with the highest anisotropy concentrations in the middle layers. This pattern diverges from the more uniformly distributed anisotropy observed in encoders. In addition, we found that the intrinsic dimension of embeddings increases in the initial phases of training, indicating an expansion into higher-dimensional space. This fact is then followed by a compression phase towards the end of training with dimensionality decrease, suggesting a refinement into more compact representations. Our results provide fresh insights to the understanding of encoders and decoders embedding properties.
This paper reveals a novel linear characteristic exclusive to transformer decoders, including models like GPT, LLaMA, OPT, BLOOM and others. We analyze embedding transformations between sequential layers, uncovering an almost perfect linear relationship (Procrustes similarity score of 0.99). However, linearity decreases when the residual component is removed, due to a consistently low transformer layer output norm. Our experiments show that pruning or linearly approximating some of the layers does not impact loss or model performance significantly. Moreover, we introduce a cosine-similarity-based regularization in our pretraining experiments on smaller models, aimed at reducing layer linearity. This regularization not only improves performance metrics on benchmarks like Tiny Stories and SuperGLUE but as well successfully decreases the linearity of the models. This study challenges the existing understanding of transformer architectures, suggesting that their operation may be more linear than previously assumed.
In this work, we examine the vulnerability of language models to universal adversarial triggers (UATs). We propose a new white-box approach to the construction of layerwise UATs (LUATs), which searches the triggers by perturbing hidden layers of a network. On the example of three transformer models and three datasets from the GLUE benchmark, we demonstrate that our method provides better transferability in a model-to-model setting with an average gain of 9.3% in the fooling rate over the baseline. Moreover, we investigate triggers transferability in the task-to-task setting. Using small subsets from the datasets similar to the target tasks for choosing a perturbed layer, we show that LUATs are more efficient than vanilla UATs by 7.1% in the fooling rate.
The embedding layers transforming input words into real vectors are the key components of deep neural networks used in natural language processing. However, when the vocabulary is large, the corresponding weight matrices can be enormous, which precludes their deployment in a limited resource setting. We introduce a novel way of parameterizing embedding layers based on the Tensor Train decomposition, which allows compressing the model significantly at the cost of a negligible drop or even a slight gain in performance. We evaluate our method on a wide range of benchmarks in natural language processing and analyze the trade-off between performance and compression ratios for a wide range of architectures, from MLPs to LSTMs and Transformers.
Skip-Gram Negative Sampling (SGNS) word embedding model, well known by its implementation in “word2vec” software, is usually optimized by stochastic gradient descent. However, the optimization of SGNS objective can be viewed as a problem of searching for a good matrix with the low-rank constraint. The most standard way to solve this type of problems is to apply Riemannian optimization framework to optimize the SGNS objective over the manifold of required low-rank matrices. In this paper, we propose an algorithm that optimizes SGNS objective using Riemannian optimization and demonstrates its superiority over popular competitors, such as the original method to train SGNS and SVD over SPPMI matrix.