{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/grokking-group-multiplication-with-cosets","title":"Grokking Group Multiplication with Cosets","arxiv_id":"2312.06581","date":"2023-12-11","proceeding":null,"authors":["Dashiell Stander","Qinan Yu","Honglu Fan","Stella Biderman"],"abstract":"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 $S_5$ and $S_6$. 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.","url_abs":"https://arxiv.org/abs/2312.06581v2","url_pdf":"https://arxiv.org/pdf/2312.06581v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"grokking-group-multiplication-with-cosets","repo_url":"https://github.com/dashstander/sn-grok","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"feedforward-network","method_name":"Feedforward Network"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2312.06581","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2312.06581"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dashstander/sn-grok","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":10,"unverified":2},"by_repo_kind":{"official":{"samples":12,"ran":10,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"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"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"a3dbc8d760de0fb0","entry":"add_fixed_to_cycle","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/groups.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/groups.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a3dbc8d760de0fb0"}},{"code_sha256_prefix":"a5e8d691fff67c68","entry":"adj_trans_decomp","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/irreps.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/irreps.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a5e8d691fff67c68"}},{"code_sha256_prefix":"077294a9dec0892e","entry":"calc_power","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/fourier.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/fourier.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"077294a9dec0892e"}},{"code_sha256_prefix":"06fa93d22a0c073a","entry":"cycle_to_one_line","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/groups.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/groups.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"06fa93d22a0c073a"}},{"code_sha256_prefix":"81bfc94d377a2d59","entry":"cycle_to_one_line","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/irreps.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/irreps.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"81bfc94d377a2d59"}},{"code_sha256_prefix":"83be00ea953442ee","entry":"dihedral_conjugacy_classes","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/dihedral_irreps.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/dihedral_irreps.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"83be00ea953442ee"}},{"code_sha256_prefix":"4a40a7258530a678","entry":"get_optimizer","repo":"dashstander/sn-grok","repo_kind":"official","path":"train_grok.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/train_grok.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4a40a7258530a678"}},{"code_sha256_prefix":"65a18a30de63b6e8","entry":"loss_fn","repo":"dashstander/sn-grok","repo_kind":"official","path":"calculate_cosets.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/calculate_cosets.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"65a18a30de63b6e8"}},{"code_sha256_prefix":"c6e17d3707e7aaf7","entry":"trans_to_one_line","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/irreps.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/irreps.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c6e17d3707e7aaf7"}},{"code_sha256_prefix":"d53845bba0da077f","entry":"union","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/cosets.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/cosets.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d53845bba0da077f"}},{"code_sha256_prefix":"b57d0d7cb168da93","entry":"slow_an_ft_1d","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/fourier.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/fourier.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b57d0d7cb168da93"}},{"code_sha256_prefix":"ee67d0f3ec236897","entry":"slow_sn_ft_1d","repo":"dashstander/sn-grok","repo_kind":"official","path":"sngrok/fourier.py","file_url":"https://github.com/dashstander/sn-grok/blob/HEAD/sngrok/fourier.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ee67d0f3ec236897"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}