# Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals

arXiv: https://arxiv.org/abs/2402.11655 · HTML: https://mlsanity.com/paper/2402.11655/

## Key facts

- Cited by 40 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 15 references that are in the map.
- First posted on arXiv 18 Feb 2024.

## Builds on (most cited first, 6 of 6)

- [How Can We Know What Language Models Know?](https://mlsanity.com/paper/1911.12543/) (2019)
- [Progress measures for grokking via mechanistic interpretability](https://mlsanity.com/paper/2301.05217/) (2023)
- [How does GPT-2 compute greater-than?: Interpreting mathematical abilities in a pre-trained language model](https://mlsanity.com/paper/2305.00586/) (2023)
- [Inverse scaling can become U-shaped](https://mlsanity.com/paper/2211.02011/) (2022)
- [Copy Suppression: Comprehensively Understanding an Attention Head](https://mlsanity.com/paper/2310.04625/) (2023)
- [Hierarchical nucleation in deep neural networks](https://mlsanity.com/paper/2007.03506/) (2020)

## Built on by (most cited first, 30 of 40)

- [A Primer on the Inner Workings of Transformer-based Language Models](https://mlsanity.com/paper/2405.00208/) (2024)
- [Attention Heads of Large Language Models: A Survey](https://mlsanity.com/paper/2409.03752/) (2024)
- [Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering](https://mlsanity.com/paper/2410.15999/) (2024)
- [Activation Scaling for Steering and Interpreting Language Models](https://mlsanity.com/paper/2410.04962/) (2024)
- [Interpreting Key Mechanisms of Factual Recall in Transformer-Based Language Models](https://mlsanity.com/paper/2403.19521/) (2024)
- [Intrinsic Test of Unlearning Using Parametric Knowledge Traces](https://mlsanity.com/paper/2406.11614/) (2024)
- [Controllable Context Sensitivity and the Knob Behind It](https://mlsanity.com/paper/2411.07404/) (2024)
- [Analysing the Residual Stream of Language Models Under Knowledge Conflicts](https://mlsanity.com/paper/2410.16090/) (2024)
- [Head Pursuit: Probing Attention Specialization in Multimodal Transformers](https://mlsanity.com/paper/2510.21518/) (2025)
- [Counterfactual Token Generation in Large Language Models](https://mlsanity.com/paper/2409.17027/) (2024)
- [Understanding and Leveraging the Expert Specialization of Context Faithfulness in Mixture-of-Experts LLMs](https://mlsanity.com/paper/2508.19594/) (2025)
- [The Atlas of In-Context Learning: How Attention Heads Shape In-Context Retrieval Augmentation](https://mlsanity.com/paper/2505.15807/) (2025)
- [When Seeing Overrides Knowing: Disentangling Knowledge Conflicts in Vision-Language Models](https://mlsanity.com/paper/2507.13868/) (2025)
- [Repetitions are not all alike: distinct mechanisms sustain repetition in language models](https://mlsanity.com/paper/2504.01100/) (2025)
- [Reasoning or Retrieval? A Study of Answer Attribution on Large Reasoning Models](https://mlsanity.com/paper/2509.24156/) (2025)
- [Generalisation First, Memorisation Second? Memorisation Localisation for Natural Language Classification Tasks](https://mlsanity.com/paper/2408.04965/) (2024)
- [A Reality Check on Context Utilisation for Retrieval-Augmented Generation](https://mlsanity.com/paper/2412.17031/) (2024)
- [Why Safeguarded Ships Run Aground? Aligned Large Language Models' Safety Mechanisms Tend to Be Anchored in The Template Region](https://mlsanity.com/paper/2502.13946/) (2025)
- [Instruction Anchor: Dissecting the Mechanistic Dynamics of Modality Arbitration](https://mlsanity.com/paper/2602.03677/) (2026)
- [Where Knowledge Collides: A Mechanistic Study of Intra-Memory Knowledge Conflict in Language Models](https://mlsanity.com/paper/2601.09445/) (2026)
- [CUB: Benchmarking Context Utilisation Techniques for Language Models](https://mlsanity.com/paper/2505.16518/) (2025)
- [Non-Interactive Symbolic-Aided Chain-of-Thought for Logical Reasoning](https://mlsanity.com/paper/2508.12425/) (2025)
- [Understanding Fact Recall in Language Models: Why Two-Stage Training Encourages Memorization but Mixed Training Teaches Knowledge](https://mlsanity.com/paper/2505.16178/) (2025)
- [Neuron-Level Analysis of Cultural Understanding in Large Language Models](https://mlsanity.com/paper/2510.08284/) (2025)
- [Where does an LLM begin computing an instruction?](https://mlsanity.com/paper/2511.10694/) (2025)
- [To Copy or Not to Copy: Copying Is Easier to Induce Than Recall](https://mlsanity.com/paper/2601.12075/) (2026)
- [Diagnosing LLM Arbitration Behavior over Pre-evidence Epistemic States in RAG-based Fact-Checking](https://mlsanity.com/paper/2606.01120/) (2026)
- [Identifying Sub-networks in Neural Networks via Functionally Similar Representations](https://mlsanity.com/paper/2410.16484/) (2024)
- [Time Course MechInterp: Analyzing the Evolution of Components and Knowledge in Large Language Models](https://mlsanity.com/paper/2506.03434/) (2025)
- [On the Generalizability of "Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals"](https://mlsanity.com/paper/2506.22977/) (2025)

## Abstract (arXiv)

Interpretability research aims to bridge the gap between empirical success and our scientific understanding of the inner workings of large language models (LLMs). However, most existing research focuses on analyzing a single mechanism, such as how models copy or recall factual knowledge. In this work, we propose a formulation of competition of mechanisms, which focuses on the interplay of multiple mechanisms instead of individual mechanisms and traces how one of them becomes dominant in the final prediction. We uncover how and where mechanisms compete within LLMs using two interpretability methods: logit inspection and attention modification. Our findings show traces of the mechanisms and their competition across various model components and reveal attention positions that effectively control the strength of certain mechanisms. Code: https://github.com/francescortu/comp-mech. Data: https://huggingface.co/datasets/francescortu/comp-mech.