Recent research has shown that factorizing large-scale language models for inference is an effective approach to improving model efficiency, significantly reducing model weights with minimal impact on performance. Interestingly, factorization can sometimes even improve accuracy by removing the noise that accumulates during training, particularly through matrix decompositions. However, recent work has primarily focused on single-matrix decompositions or lower precision techniques, which may fail to fully capture structural patterns. To address these limitations, we introduce TRAWL (Tensor Reduced and Approximated Weights for Large Language Models), a technique that applies tensor decomposition across multiple weight matrices to effectively denoise LLMs by capturing both global and local structural patterns. Our experiments show that TRAWL improves the model performance by up to 16% over baseline models on benchmark datasets, without requiring additional data, training, or fine-tuning.