@article{suppressingfinallayerhiddenstatejumpsint, title = {Suppressing Final Layer Hidden State Jumps in Transformer Pretraining}, author = {Keigo Shibata and Kazuki Yano and Ryosuke Takahashi and Jaesung Lee and Wataru Ikeda and Jun Suzuki}, year = {2026}, eprint = {2601.18302}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2601.18302}, }