{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:XXH3CVNX3D6R3Z62WSLZPACUI2","short_pith_number":"pith:XXH3CVNX","schema_version":"1.0","canonical_sha256":"bdcfb155b7d8fd1de7dab497978054469d4c65e04b4e5908f2c20294729d69e9","source":{"kind":"arxiv","id":"2605.21486","version":1},"attestation_state":"computed","paper":{"title":"Quantifying Hyperparameter Transfer and the Importance of Embedding Layer Learning Rate","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dayal Singh Kalra, Maissam Barkeshli","submitted_at":"2026-05-20T17:59:40Z","abstract_excerpt":"Hyperparameter transfer allows extrapolating optimal optimization hyperparameters from small to large scales, making it critical for training large language models (LLMs). This is done either by fitting a scaling law to the hyperparameters or by a judicious choice of parameterization, such as Maximal Update ($\\mu$P), that renders optimal hyperparameters approximately scale invariant. In this paper, we first develop a framework to quantify hyperparameter transfer through three metrics: (1) the quality of the scaling law fit, (2) the robustness to extrapolation errors, and (3) the asymptotic los"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.21486","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T17:59:40Z","cross_cats_sorted":["cond-mat.dis-nn","cs.AI","stat.ML"],"title_canon_sha256":"2ac32a55a31c1da37b9f56bafc9d0dc399c5d7a20291e6d9c802c1c1849997b5","abstract_canon_sha256":"0bca2bc7275628cc8f1be9fc5e56db3b0d21c92e6c2fb743ba225632fb394a49"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T02:05:39.558973Z","signature_b64":"nokNyOzfWtE3Go2dhqDlcpct4X4kZpaR1Tm0Kt4tyzZArQiRBFuhEklEnMjD5sjXM5CeXNVo/xQAcBLpTxh1BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bdcfb155b7d8fd1de7dab497978054469d4c65e04b4e5908f2c20294729d69e9","last_reissued_at":"2026-05-21T02:05:39.558199Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T02:05:39.558199Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Quantifying Hyperparameter Transfer and the Importance of Embedding Layer Learning Rate","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dayal Singh Kalra, Maissam Barkeshli","submitted_at":"2026-05-20T17:59:40Z","abstract_excerpt":"Hyperparameter transfer allows extrapolating optimal optimization hyperparameters from small to large scales, making it critical for training large language models (LLMs). This is done either by fitting a scaling law to the hyperparameters or by a judicious choice of parameterization, such as Maximal Update ($\\mu$P), that renders optimal hyperparameters approximately scale invariant. In this paper, we first develop a framework to quantify hyperparameter transfer through three metrics: (1) the quality of the scaling law fit, (2) the robustness to extrapolation errors, and (3) the asymptotic los"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21486","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.21486/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.21486","created_at":"2026-05-21T02:05:39.558323+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21486v1","created_at":"2026-05-21T02:05:39.558323+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21486","created_at":"2026-05-21T02:05:39.558323+00:00"},{"alias_kind":"pith_short_12","alias_value":"XXH3CVNX3D6R","created_at":"2026-05-21T02:05:39.558323+00:00"},{"alias_kind":"pith_short_16","alias_value":"XXH3CVNX3D6R3Z62","created_at":"2026-05-21T02:05:39.558323+00:00"},{"alias_kind":"pith_short_8","alias_value":"XXH3CVNX","created_at":"2026-05-21T02:05:39.558323+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XXH3CVNX3D6R3Z62WSLZPACUI2","json":"https://pith.science/pith/XXH3CVNX3D6R3Z62WSLZPACUI2.json","graph_json":"https://pith.science/api/pith-number/XXH3CVNX3D6R3Z62WSLZPACUI2/graph.json","events_json":"https://pith.science/api/pith-number/XXH3CVNX3D6R3Z62WSLZPACUI2/events.json","paper":"https://pith.science/paper/XXH3CVNX"},"agent_actions":{"view_html":"https://pith.science/pith/XXH3CVNX3D6R3Z62WSLZPACUI2","download_json":"https://pith.science/pith/XXH3CVNX3D6R3Z62WSLZPACUI2.json","view_paper":"https://pith.science/paper/XXH3CVNX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21486&json=true","fetch_graph":"https://pith.science/api/pith-number/XXH3CVNX3D6R3Z62WSLZPACUI2/graph.json","fetch_events":"https://pith.science/api/pith-number/XXH3CVNX3D6R3Z62WSLZPACUI2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XXH3CVNX3D6R3Z62WSLZPACUI2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XXH3CVNX3D6R3Z62WSLZPACUI2/action/storage_attestation","attest_author":"https://pith.science/pith/XXH3CVNX3D6R3Z62WSLZPACUI2/action/author_attestation","sign_citation":"https://pith.science/pith/XXH3CVNX3D6R3Z62WSLZPACUI2/action/citation_signature","submit_replication":"https://pith.science/pith/XXH3CVNX3D6R3Z62WSLZPACUI2/action/replication_record"}},"created_at":"2026-05-21T02:05:39.558323+00:00","updated_at":"2026-05-21T02:05:39.558323+00:00"}