{"id":"https://openalex.org/W7213498564","doi":"https://doi.org/10.5281/zenodo.22813297","title":"The Carried Reference Tunes a Switch\u2013Multiplier Continuum in Gated (GLU) MLPs","display_name":"The Carried Reference Tunes a Switch\u2013Multiplier Continuum in Gated (GLU) MLPs","publication_year":2026,"publication_date":"2026-09-17","ids":{"openalex":"https://openalex.org/W7213498564","doi":"https://doi.org/10.5281/zenodo.22813297"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.22813297","doi":"https://doi.org/10.5281/zenodo.22813297","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.22813297","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.22813297","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5017808661","display_name":"Olli Tuomi","orcid":"https://orcid.org/0009-0006-2042-1576","observed_orcids":["https://orcid.org/0009-0006-2042-1576"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Tuomi, Olli","raw_affiliation_strings":["Evident Solutions Oy"],"raw_orcid":"https://orcid.org/0009-0006-2042-1576","affiliations":[{"raw_affiliation_string":"Evident Solutions Oy","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5017808661"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.3544999957084656,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.3544999957084656,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.31859999895095825,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.11069999635219574,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/swiglu","display_name":"SwiGLU","score":0.8040000200271606},{"id":"https://openalex.org/keywords/mixture-of-experts","display_name":"mixture of experts","score":0.550000011920929}],"concepts":[{"id":"https://openalex.org/C158622935","wikidata":"https://www.wikidata.org/wiki/Q660848","display_name":"Nonlinear system","level":2,"score":0.4715999960899353},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4652000069618225},{"id":"https://openalex.org/C135598885","wikidata":"https://www.wikidata.org/wiki/Q1366302","display_name":"Row","level":2,"score":0.46149998903274536},{"id":"https://openalex.org/C124584101","wikidata":"https://www.wikidata.org/wiki/Q1053266","display_name":"Multiplier (economics)","level":2,"score":0.45579999685287476},{"id":"https://openalex.org/C205203396","wikidata":"https://www.wikidata.org/wiki/Q612143","display_name":"Bilinear interpolation","level":2,"score":0.4205999970436096},{"id":"https://openalex.org/C184720557","wikidata":"https://www.wikidata.org/wiki/Q7825049","display_name":"Topology (electrical circuits)","level":2,"score":0.41940000653266907},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.40209999680519104},{"id":"https://openalex.org/C186661526","wikidata":"https://www.wikidata.org/wiki/Q13647261","display_name":"Sine","level":2,"score":0.3727000057697296},{"id":"https://openalex.org/C131584629","wikidata":"https://www.wikidata.org/wiki/Q4308705","display_name":"Coupling (piping)","level":2,"score":0.3682999908924103},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.358599990606308},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.3569999933242798},{"id":"https://openalex.org/C178009071","wikidata":"https://www.wikidata.org/wiki/Q93344","display_name":"Trigonometric functions","level":2,"score":0.3384999930858612},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.33149999380111694},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.3167000114917755},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3118000030517578},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.28999999165534973},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.2858999967575073},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2750000059604645},{"id":"https://openalex.org/C131017901","wikidata":"https://www.wikidata.org/wiki/Q170451","display_name":"Logic gate","level":2,"score":0.2745000123977661},{"id":"https://openalex.org/C194544171","wikidata":"https://www.wikidata.org/wiki/Q21105679","display_name":"Gating","level":2,"score":0.2590000033378601},{"id":"https://openalex.org/C3019835501","wikidata":"https://www.wikidata.org/wiki/Q1310130","display_name":"Single shot","level":2,"score":0.2581000030040741},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.25519999861717224},{"id":"https://openalex.org/C2992734406","wikidata":"https://www.wikidata.org/wiki/Q413267","display_name":"One shot","level":2,"score":0.2533999979496002}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.22813297","doi":"https://doi.org/10.5281/zenodo.22813297","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.22813297","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},{"id":"doi:10.5281/zenodo.22813298","doi":"https://doi.org/10.5281/zenodo.22813298","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.22813298","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.5281/zenodo.22813297","doi":"https://doi.org/10.5281/zenodo.22813297","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.22813297","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7170077615","https://openalex.org/W3006439205","https://openalex.org/W7163171047","https://openalex.org/W7170044523","https://openalex.org/W7213776806","https://openalex.org/W7138008801","https://openalex.org/W7167578150","https://openalex.org/W7214314845","https://openalex.org/W4416512505","https://openalex.org/W7133828067"],"abstract_inverted_index":{"A":[0],"prior":[1],"paper":[2],"showed":[3],"that":[4,102,161],"a":[5,14,19,34,41,53,57,68,72,86,147,187,204],"trained":[6,105],"transformer":[7],"implements":[8],"its":[9,90],"MLP":[10],"gate":[11,27,39,46],"thresholds":[12],"as":[13,67,186],"couplingw":[15],"\u00b7":[16],"bL":[17,76,135],"to":[18,32,62,75,82,123,200],"carried":[20],"reference":[21],"direction":[22],"bL,":[23],"the":[24,38,79,99,113,120,124,130,138,143,157,166,189,198,201],"corpus-mean":[25],"layer-normalised":[26],"input.":[28],"We":[29,178],"use":[30],"thatreference":[31],"give":[33],"functional":[35],"reading":[36],"of":[37,89,203],"in":[40,142],"gated":[42,167],"(GLU)":[43],"MLP.":[44],"Each":[45],"unit\u2019s":[47],"coupling":[48,121],"tobL":[49,97],"places":[50],"it":[51],"on":[52],"continuous":[54],"switch\u2013multiplier":[55],"spectrum:":[56],"unit":[58,73],"strongly":[59],"(negatively)":[60],"coupled":[61],"bLrests":[63],"closed":[64],"and":[65,104,140,160,175,196,210],"behaves":[66],"thresholded":[69],"switch,":[70],"while":[71],"orthogonal":[74],"rests":[77],"at":[78,136],"activation\u2019sknee":[80],"and,":[81],"first":[83],"order,":[84],"computes":[85],"bilinear":[87],"product":[88],"two":[91],"branch":[92],"reads.":[93,177],"The":[94,206],"per-unit":[95],"cosine":[96],"is":[98,154,192,208],"coordinate":[100],"along":[101],"spectrum,":[103],"SwiGLU":[106,128],"allocates":[107],"units":[108],"continuously":[109],"acrossit,":[110],"shifting":[111],"toward":[112],"multiplier":[114],"end":[115],"with":[116,134,172],"depth.":[117],"Supporting":[118],"this,":[119],"localises":[122],"gatebranch":[125],"(across":[126],"four":[127,163],"models":[129],"gate_proj":[131],"rows":[132],"align":[133],"4\u201333\u00d7":[137],"up_projrows)":[139],"lives":[141],"reference\u2019s":[144],"high-variance":[145],"subspace;":[146],"bit-identical-init":[148],"training":[149],"ladder":[150],"shows":[151],"theelementwise":[152],"nonlinearity":[153],"what":[155],"creates":[156],"switch":[158],"end,":[159],"among":[162],"feed-forward":[164],"variantsonly":[165],"one":[168],"spans":[169],"both":[170],"ends":[171],"independent":[173],"key":[174],"value":[176],"suggest":[179],"this":[180],"coverage,rather":[181],"than":[182],"any":[183],"single":[184],"mode,":[185],"reason":[188],"GLU":[190],"family":[191],"an":[193],"effective":[194],"default,":[195],"note":[197],"readingextends":[199],"experts":[202],"mixture-of-experts.":[205],"account":[207],"geometric":[209],"makes":[211],"no":[212],"quality":[213],"claim":[214]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-18T00:00:00"}
