Papers › Grokking Group Multiplication with Cosets

Grokking Group Multiplication with Cosets

11 Dec 2023arXiv:2312.06581archive 2025-07-28

Dashiell Stander, Qinan Yu, Honglu Fan, Stella Biderman

The complex and unpredictable nature of deep neural networks prevents their safe use in many high-stakes applications. There have been many techniques developed to interpret deep neural networks, but all have substantial limitations. Algorithmic tasks have proven to be a fruitful test ground for interpreting a neural network end-to-end. Building on previous work, we completely reverse engineer fully connected one-hidden layer networks that have ``grokked'' the arithmetic of the permutation groups S₅ and S₆. The models discover the true subgroup structure of the full group and converge on neural circuits that decompose the group arithmetic using the permutation group's subgroups. We relate how we reverse engineered the model's mechanisms and confirmed our theory was a faithful description of the circuit's functionality. We also draw attention to current challenges in conducting interpretability research by comparing our work to Chughtai et al. [4] which alleges to find a different algorithm for this same problem.

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2312.06581")

Code

Syntology Ran 10 of 12 code samples harvested from 1 repository linked to this paper; 2 have no recorded run. Of those that ran: 10 ran with no contract checked.

By repository: official repository: 12 samples from 1 repository, 10 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

dashstander/sn-grok officialmentioned in paperpytorchMIT 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

12 samples harvested; 10 ran; 0 honoured the contract we drafted; 2 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

10ran
2unverified

Licence: 0 of the 12 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 dashstander/sn-grok. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

add_fixed_to_cycle dashstander/sn-grok/sngrok/groups.py official repository ran MIT (permissive) · a3dbc8d760de0fb0 · report
adj_trans_decomp dashstander/sn-grok/sngrok/irreps.py official repository ran fingerprinted MIT (permissive) · a5e8d691fff67c68 · report
calc_power dashstander/sn-grok/sngrok/fourier.py official repository ran MIT (permissive) · 077294a9dec0892e · report
cycle_to_one_line dashstander/sn-grok/sngrok/groups.py official repository ran MIT (permissive) · 06fa93d22a0c073a · report
cycle_to_one_line dashstander/sn-grok/sngrok/irreps.py official repository ran MIT (permissive) · 81bfc94d377a2d59 · report
dihedral_conjugacy_classes dashstander/sn-grok/sngrok/dihedral_irreps.py official repository ran MIT (permissive) · 83be00ea953442ee · report
get_optimizer dashstander/sn-grok/train_grok.py official repository ran MIT (permissive) · 4a40a7258530a678 · report
loss_fn dashstander/sn-grok/calculate_cosets.py official repository ran MIT (permissive) · 65a18a30de63b6e8 · report
trans_to_one_line dashstander/sn-grok/sngrok/irreps.py official repository ran MIT (permissive) · c6e17d3707e7aaf7 · report
union dashstander/sn-grok/sngrok/cosets.py official repository ran MIT (permissive) · d53845bba0da077f · report
slow_an_ft_1d dashstander/sn-grok/sngrok/fourier.py official repository unverified MIT (permissive) · b57d0d7cb168da93 · report
slow_sn_ft_1d dashstander/sn-grok/sngrok/fourier.py official repository unverified MIT (permissive) · ee67d0f3ec236897 · report

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Dense ConnectionsFeedforward Network

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