# Transformers Don't Need LayerNorm at Inference Time: Scaling LayerNorm Removal to GPT-2 XL and the Implications for Mechanistic Interpretability

arXiv: https://arxiv.org/abs/2507.02559 · HTML: https://mlsanity.com/paper/2507.02559/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge and representations inside LLMs, one of 13 directions in Mechanistic interpretability.
- Builds on 22 of its 34 references that are in the map.
- First posted on arXiv 3 Jul 2025.

## Placed in

- Mechanistic interpretability / [Knowledge and representations inside LLMs](https://mlsanity.com/mechinterp/knowledge-and-representations-inside-llms/)

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Root Mean Square Layer Normalization](https://mlsanity.com/paper/1910.07467/) (2019)
- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models](https://mlsanity.com/paper/2403.19647/) (2024)
- [Towards Best Practices of Activation Patching in Language Models: Metrics and Methods](https://mlsanity.com/paper/2309.16042/) (2023)
- [Direct and Indirect Effects](https://mlsanity.com/paper/1301.2300/) (2013)
- [How to use and interpret activation patching](https://mlsanity.com/paper/2404.15255/) (2024)
- [Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representations](https://mlsanity.com/paper/2303.02536/) (2023)
- [Why do LLMs attend to the first token?](https://mlsanity.com/paper/2504.02732/) (2025)
- [Universal Neurons in GPT2 Language Models](https://mlsanity.com/paper/2401.12181/) (2024)
- [Copy Suppression: Comprehensively Understanding an Attention Head](https://mlsanity.com/paper/2310.04625/) (2023)
- [Confidence Regulation Neurons in Language Models](https://mlsanity.com/paper/2406.16254/) (2024)
- [VISIT: Visualizing and Interpreting the Semantic Information Flow of Transformers](https://mlsanity.com/paper/2305.13417/) (2023)
- [Interpreting the Repeated Token Phenomenon in Large Language Models](https://mlsanity.com/paper/2503.08908/) (2025)
- [Understanding the Failure of Batch Normalization for Transformers in NLP](https://mlsanity.com/paper/2210.05153/) (2022)
- [You can remove GPT2's LayerNorm by fine-tuning](https://mlsanity.com/paper/2409.13710/) (2024)
- [Jacobian Sparse Autoencoders: Sparsify Computations, Not Just Activations](https://mlsanity.com/paper/2502.18147/) (2025)
- [The Local Interaction Basis: Identifying Computationally-Relevant and Sparsely Interacting Features in Neural Networks](https://mlsanity.com/paper/2405.10928/) (2024)

## Built on by (most cited first, 7 of 7)

- [Gated Normalization Removal and Scale Anchoring in Pre-Norm Transformers](https://mlsanity.com/paper/2602.10408/) (2026)
- [Key and Value Weights Are Probably All You Need: On the Necessity of the Query, Key, Value weight Triplet in Self-Attention Transformers](https://mlsanity.com/paper/2510.23912/) (2025)
- [Towards Verifiable Transformers: Solver-Checkable Circuit Explanations](https://mlsanity.com/paper/2605.24033/) (2026)
- [Robust Harmful Features Under Jailbreak Attacks: Mechanistic Evidence from Attention Head Specialization in Large Language Models](https://mlsanity.com/paper/2606.28153/) (2026)
- [Selective Neuron Amplification in Transformer Language Models](https://mlsanity.com/paper/2604.07098/) (2026)
- [Can an MLP Absorb Its Own Skip Connection Exactly?](https://mlsanity.com/paper/2604.23705/) (2026)
- [Deep Noir: Autonomous Steering Discovery via Architectural Chronometry in Transformer Models](https://mlsanity.com/paper/2609.20722/) (2026)

## Abstract (arXiv)

Layer-wise normalization (LN) is an essential component of virtually all transformer-based large language models. While its effects on training stability are well documented, its role at inference time is poorly understood. Additionally, LN layers hinder mechanistic interpretability by introducing additional nonlinearities and increasing the interconnectedness of individual model components. Here, we show that all LN layers can be removed from every GPT-2 model with only a small increase in validation loss (e.g. +0.03 cross-entropy loss for GPT-2 XL). Thus, LN cannot play a substantial role in language modeling. We find that the amount of fine-tuning data needed for LN removal grows sublinearly with model parameters, suggesting scaling to larger models is feasible. We release a suite of LN-free GPT-2 models on Hugging Face. Furthermore, we test interpretability techniques on LN-free models. Direct logit attribution now gives the exact direct effect of individual components, while the accuracy of attribution patching does not significantly improve. We also confirm that GPT-2's "confidence neurons" are inactive in the LN-free models. Our work clarifies the role of LN layers in language modeling, showing that GPT-2-class models can function without LN layers. We hope that our LN-free analogs of the GPT-2 family of models will enable more precise interpretability research and improve our understanding of language models.