@article{doesthelmheadcreateaharmfulgradientbottl, title = {Does the LM Head Create a Harmful Gradient Bottleneck? A Causal Test}, author = {Anand Murugan}, year = {2026}, eprint = {2608.16671}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2608.16671}, }