{"id":"https://openalex.org/W7225423630","doi":"https://doi.org/10.5281/zenodo.23992384","title":"From Fixed Points to Scaling Laws: A Parameter-Free Prediction of Grokking Phase Transitions in Large Language Models Using the Functional Renormalization Group","display_name":"From Fixed Points to Scaling Laws: A Parameter-Free Prediction of Grokking Phase Transitions in Large Language Models Using the Functional Renormalization Group","publication_year":2026,"publication_date":"2026-11-06","ids":{"openalex":"https://openalex.org/W7225423630","doi":"https://doi.org/10.5281/zenodo.23992384"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.23992384","doi":"https://doi.org/10.5281/zenodo.23992384","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23992384","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.23992384","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124350434","display_name":"Alis Hasi\u0107","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Hasi\u0107, Alis","raw_affiliation_strings":["Independent Researcher"],"raw_orcid":"https://orcid.org/0009-0008-5799-3345","affiliations":[{"raw_affiliation_string":"Independent Researcher","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5124350434"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.25589999556541443,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.25589999556541443,"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/T10591","display_name":"Theoretical and Computational Physics","score":0.20800000429153442,"subfield":{"id":"https://openalex.org/subfields/3104","display_name":"Condensed Matter Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.20239999890327454,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/grokking","display_name":"grokking","score":0.7760000228881836},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.9440000057220459},{"id":"https://openalex.org/keywords/functional-renormalization-group","display_name":"functional renormalization group","score":0.9419999718666077},{"id":"https://openalex.org/keywords/fisher-information-metric","display_name":"Fisher information metric","score":0.7369999885559082},{"id":"https://openalex.org/keywords/statistical-field-theory","display_name":"statistical field theory","score":0.878000020980835},{"id":"https://openalex.org/keywords/overparameterized-neural-networks","display_name":"overparameterized neural networks","score":0.7860000133514404},{"id":"https://openalex.org/keywords/training-dynamics","display_name":"training dynamics","score":0.6959999799728394},{"id":"https://openalex.org/keywords/infrared-fixed-point","display_name":"infrared fixed point","score":0.8069999814033508},{"id":"https://openalex.org/keywords/critical-exponents","display_name":"critical exponents","score":0.9480000138282776},{"id":"https://openalex.org/keywords/non-equilibrium-steady-state","display_name":"nonequilibrium steady states","score":0.8180000185966492},{"id":"https://openalex.org/keywords/spectral-gap","display_name":"spectral gap","score":0.781000018119812},{"id":"https://openalex.org/keywords/scaling-laws","display_name":"scaling laws","score":0.6710000038146973},{"id":"https://openalex.org/keywords/finite-size-corrections","display_name":"finite-size corrections","score":0.734000027179718},{"id":"https://openalex.org/keywords/data-collapse","display_name":"data collapse","score":0.5870000123977661},{"id":"https://openalex.org/keywords/ai-safety","display_name":"AI safety","score":0.8370000123977661}],"concepts":[{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5985000133514404},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.5860999822616577},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5698000192642212},{"id":"https://openalex.org/C68532491","wikidata":"https://www.wikidata.org/wiki/Q1203669","display_name":"Renormalization group","level":2,"score":0.5669000148773193},{"id":"https://openalex.org/C61445026","wikidata":"https://www.wikidata.org/wiki/Q217608","display_name":"Fixed point","level":2,"score":0.5569999814033508},{"id":"https://openalex.org/C149288129","wikidata":"https://www.wikidata.org/wiki/Q185357","display_name":"Phase transition","level":2,"score":0.43299999833106995},{"id":"https://openalex.org/C203616005","wikidata":"https://www.wikidata.org/wiki/Q620495","display_name":"Hessian matrix","level":2,"score":0.4092000126838684},{"id":"https://openalex.org/C183992945","wikidata":"https://www.wikidata.org/wiki/Q2495574","display_name":"Universality (dynamical systems)","level":2,"score":0.40689998865127563},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.37549999356269836},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.3711000084877014},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.35359999537467957},{"id":"https://openalex.org/C2781181686","wikidata":"https://www.wikidata.org/wiki/Q4226068","display_name":"Coherence (philosophical gambling strategy)","level":2,"score":0.32420000433921814},{"id":"https://openalex.org/C72169020","wikidata":"https://www.wikidata.org/wiki/Q194404","display_name":"Monotonic function","level":2,"score":0.32199999690055847},{"id":"https://openalex.org/C194807282","wikidata":"https://www.wikidata.org/wiki/Q900634","display_name":"Critical phenomena","level":3,"score":0.31850001215934753},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.28450000286102295},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.2802000045776367},{"id":"https://openalex.org/C2778114796","wikidata":"https://www.wikidata.org/wiki/Q7575194","display_name":"Spectral gap","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.2752000093460083},{"id":"https://openalex.org/C17231256","wikidata":"https://www.wikidata.org/wiki/Q5156540","display_name":"Completeness (order theory)","level":2,"score":0.2743000090122223},{"id":"https://openalex.org/C202213908","wikidata":"https://www.wikidata.org/wiki/Q626011","display_name":"Mean field theory","level":2,"score":0.2689000070095062},{"id":"https://openalex.org/C207282930","wikidata":"https://www.wikidata.org/wiki/Q5380811","display_name":"Entropy production","level":2,"score":0.2687000036239624},{"id":"https://openalex.org/C60640748","wikidata":"https://www.wikidata.org/wiki/Q2337858","display_name":"Lyapunov function","level":3,"score":0.26179999113082886},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.26170000433921814},{"id":"https://openalex.org/C39927690","wikidata":"https://www.wikidata.org/wiki/Q11197","display_name":"Logarithm","level":2,"score":0.2531999945640564}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.23992384","doi":"https://doi.org/10.5281/zenodo.23992384","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23992384","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.23992385","doi":"https://doi.org/10.5281/zenodo.23992385","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23992385","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.23992384","doi":"https://doi.org/10.5281/zenodo.23992384","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23992384","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/W7172305105","https://openalex.org/W7143398690","https://openalex.org/W7214537689","https://openalex.org/W4415201014","https://openalex.org/W4406890872","https://openalex.org/W7162298793","https://openalex.org/W7221978891","https://openalex.org/W7128504444","https://openalex.org/W7201980175","https://openalex.org/W7130574421"],"abstract_inverted_index":{"The":[0],"empirical":[1],"success":[2],"of":[3,18,35,52,90,123,132],"Large":[4],"Language":[5],"Models":[6],"(LLMs)":[7],"is":[8,104],"currently":[9],"governed":[10],"by":[11,47,119],"phenomenological":[12],"scaling":[13,83,141],"laws":[14,84],"and":[15,33,157,169],"heuristic":[16],"observations":[17],"phase":[19],"transitions,":[20],"such":[21],"as":[22,56,87],"\"Grokking\".":[23],"However,":[24],"a":[25,57,91,110,162],"rigorous,":[26],"first-principles":[27],"mathematical":[28],"framework":[29],"guaranteeing":[30],"the":[31,49,62,70,76,101,120,124,130,133,151],"stability":[32],"universality":[34],"these":[36],"dynamics":[37,51],"remains":[38],"absent.":[39],"In":[40],"this":[41,45],"work,":[42],"we":[43,98,136,144],"bridge":[44],"gap":[46],"formulating":[48],"training":[50],"overparameterized":[53],"neural":[54],"networks":[55],"statistical":[58],"field":[59],"theory":[60],"on":[61],"Fisher":[63],"Information":[64],"Metric":[65],"(FIM)":[66],"manifold.":[67],"We":[68],"derive":[69],"exact":[71],"Wetterich":[72],"flow":[73],"equation":[74],"for":[75,166],"effective":[77],"average":[78],"loss,":[79],"demonstrating":[80],"that":[81,100],"macroscopic":[82],"emerge":[85],"parameter-free":[86],"critical":[88],"exponents":[89],"non-trivial":[92],"infrared":[93],"(IR)":[94],"fixed":[95],"point.":[96],"Furthermore,":[97],"prove":[99],"Grokking":[102],"transition":[103],"not":[105],"an":[106],"optimization":[107],"artifact,":[108],"but":[109],"strict":[111],"Non-Equilibrium":[112],"Steady":[113],"State":[114],"(NESS)":[115],"Lyapunov":[116],"attractor,":[117],"driven":[118],"discontinuous":[121],"collapse":[122],"Hessian's":[125],"spectral":[126,153],"entropy.":[127],"By":[128],"establishing":[129],"completeness":[131],"O(\u2202\u00b2)":[134],"truncation,":[135],"quantify":[137],"finite-size":[138],"corrections":[139],"to":[140],"laws.":[142],"Finally,":[143],"provide":[145],"three":[146],"concrete,":[147],"falsifiable":[148],"predictions":[149],"regarding":[150],"Hessian":[152],"gap,":[154],"entropy":[155],"collapse,":[156,160],"RG":[158],"data":[159],"offering":[161],"mathematically":[163],"closed":[164],"foundation":[165],"AI":[167],"safety":[168],"core":[170],"mechanism":[171],"research.":[172]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-11-07T00:00:00"}