Papers › Emergent Modularity in Pre-trained Transformers

Emergent Modularity in Pre-trained Transformers

28 May 2023arXiv:2305.18390archive 2025-07-28

Zhengyan Zhang, Zhiyuan Zeng, Yankai Lin, Chaojun Xiao, Xiaozhi Wang, Xu Han, Zhiyuan Liu, Ruobing Xie, Maosong Sun, Jie zhou

This work examines the presence of modularity in pre-trained Transformers, a feature commonly found in human brains and thought to be vital for general intelligence. In analogy to human brains, we consider two main characteristics of modularity: (1) functional specialization of neurons: we evaluate whether each neuron is mainly specialized in a certain function, and find that the answer is yes. (2) function-based neuron grouping: we explore finding a structure that groups neurons into modules by function, and each module works for its corresponding function. Given the enormous amount of possible structures, we focus on Mixture-of-Experts as a promising candidate, which partitions neurons into experts and usually activates different experts for different inputs. Experimental results show that there are functional experts, where clustered are the neurons specialized in a certain function. Moreover, perturbing the activations of functional experts significantly affects the corresponding function. Finally, we study how modularity emerges during pre-training, and find that the modular structure is stabilized at the early stage, which is faster than neuron stabilization. It suggests that Transformers first construct the modular structure and then learn fine-grained neuron functions. Our code and data are available at https://github.com/THUNLP/modularity-analysis.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 10 code samples harvested from 1 repository linked to this paper: 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 10 samples from 1 repository, 7 ran (7 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

thunlp/modularity-analysis officialmentioned in papermentioned on GitHubpytorchMITSyntology: 10 samples, 7 ran (0 instrument-failed) · 3 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

10 samples harvested: 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

7ran
3unverified

Licence: 0 of the 10 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from thunlp/modularity-analysis. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

thunlp/modularity-analysis official repository 10 samples, 7 ran (0 instrument-failed) · 3 unverified
binom_hypothesis_testing FunctionalExpert/HypothesisTesting.py ran MIT (permissive) · 5519a08ebdb9ff1f · report fixture: signature countings:Tensor/1/int · returned tuple · level recorded 2026-09-26
get_TopK FunctionalExpert/TopK_neurons.py ran MIT (permissive) · 8306a018292a7872 · report fixture: signature ap:Tensor/1/float · returned list · level recorded 2026-09-26
get_counting FunctionalExpert/HypothesisTesting.py ran MIT (permissive) · e0168a836e0e2061 · report fixture: signature ap:list/1/int, expert_belong:dict/0/? · returned Tensor · level recorded 2026-09-26
get_counting FunctionalExpert/HypothesisTesting4RandomPartitioning.py ran MIT (permissive) · 6eafb817b3f2461f · report fixture: signature ap:list/1/int, expert_belong:list/1/int · returned Tensor · level recorded 2026-09-26
sim_topk NeuronFunctionalSpecialization/Figure1Right_maker.py ran MIT (permissive) · 39eb7a4fd06a4d3b · report fixture: signature a:Tensor/1/float, b:Tensor/1/float, k:int/0/int · returned float · level recorded 2026-09-26
sum_tool FunctionalExpert/Table1and4_maker.py ran MIT (permissive) · ab802339bdf17ca9 · report fixture: signature data:dict/0/?, index:int/0/int · returned float · level recorded 2026-09-26
task_CalAP NeuronPredictivity/ap_calculation.py ran MIT (permissive) · e2262290cd2baca5 · report fixture: signature dataset:dict/0/?, task:str/0/str, acts:dict/0/? · returned tuple · level recorded 2026-09-26
hypergeom_hypothesis_testing FunctionalExpert/HypothesisTesting.py unverified MIT (permissive) · 8e5860a7cc71c303 · report
knowledge_CalAP NeuronPredictivity/ap_calculation.py unverified MIT (permissive) · b9209b6e9d6f65bd · report
semantic_CalAP NeuronPredictivity/ap_calculation.py unverified MIT (permissive) · b5e5c9b52368bd08 · report

Tasks

Mixture-of-Experts

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Focus

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections