# Tiny Brains, Giant Impact: Uncovering the Keystone Neurons of LLM with Just a Few Prompts

arXiv: https://arxiv.org/abs/2605.24846 · HTML: https://mlsanity.com/paper/2605.24846/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 35 references that are in the map.
- First posted on arXiv 24 May 2026.

## Builds on (most cited first, 23 of 23)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks](https://mlsanity.com/paper/1803.03635/) (2018)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [OlympiadBench: A Challenging Benchmark for Promoting AGI with Olympiad-Level Bilingual Multimodal Scientific Problems](https://mlsanity.com/paper/2402.14008/) (2024)
- [Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus](https://mlsanity.com/paper/2104.08758/) (2021)
- [The Lottery Ticket Hypothesis for Pre-trained BERT Networks](https://mlsanity.com/paper/2007.12223/) (2020)
- [Recurrent Independent Mechanisms](https://mlsanity.com/paper/1909.10893/) (2019)
- [DEMix Layers: Disentangling Domains for Modular Language Modeling](https://mlsanity.com/paper/2108.05036/) (2021)
- [Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models](https://mlsanity.com/paper/2402.16438/) (2024)
- [Are Neural Nets Modular? Inspecting Functional Modularity Through Differentiable Weight Masks](https://mlsanity.com/paper/2010.02066/) (2020)
- [Interpreting Arithmetic Mechanism in Large Language Models through Comparative Neuron Analysis](https://mlsanity.com/paper/2409.14144/) (2024)
- [The Emergence of Abstract Thought in Large Language Models Beyond Any Language](https://mlsanity.com/paper/2506.09890/) (2025)
- [Unlocking Emergent Modularity in Large Language Models](https://mlsanity.com/paper/2310.10908/) (2023)
- [Not All Parameters Are Created Equal: Smart Isolation Boosts Fine-Tuning Performance](https://mlsanity.com/paper/2508.21741/) (2025)
- [Mechanistic Understanding of Language Models in Syntactic Code Completion](https://mlsanity.com/paper/2502.18499/) (2025)

## Built on by (most cited first, 2 of 2)

- [AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning](https://mlsanity.com/paper/2608.05987/) (2026)
- [Travel Back To Move Forward: Retrospective Exploration and Evolving for Long Horizon Agents](https://mlsanity.com/paper/2610.13058/) (2026)

## Abstract (arXiv)

Large language models (LLMs) display strong comprehensive abilities, yet the internal mechanisms that support these behaviors remain insufficiently understood. In this work, we show that across a wide range of open-weight Transformers, a subset of neurons remains consistently highly activated during inference across tasks of multiple capability dimensions. By probing along the cross-task activation strength, an extremely sparse subset is isolated, whose removal causes a collapse in model behavior, which we term keystone neurons. Our analysis reveals that keystone neurons are a stable and intrinsic neuron subset of the model that is largely established during pretraining. The parameters associated with these neurons are tightly calibrated during the training process, and their precise values are critical for the capabilities of the model. Building on these insights, we propose a supervised fine-tuning approach that updates only keystone neurons, achieving task gains comparable to or even better than full-parameter fine-tuning while better preserving performance in other capability dimensions, despite modifying a much smaller number of parameters.