{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:J45IOFINV7A76EEFM6UDNWXFY4","short_pith_number":"pith:J45IOFIN","schema_version":"1.0","canonical_sha256":"4f3a87150dafc1ff108567a836dae5c71d7905070724876d69de1de05d5326f6","source":{"kind":"arxiv","id":"2202.06387","version":2},"attestation_state":"computed","paper":{"title":"Scaling Laws Under the Microscope: Predicting Transformer Performance from Small Scale Experiments","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NA","math.NA"],"primary_cat":"cs.CL","authors_text":"Jonathan Berant, Maor Ivgi, Yair Carmon","submitted_at":"2022-02-13T19:13:00Z","abstract_excerpt":"Neural scaling laws define a predictable relationship between a model's parameter count and its performance after training in the form of a power law. However, most research to date has not explicitly investigated whether scaling laws can be used to accelerate model development. In this work, we perform such an empirical investigation across a wide range of language understanding tasks, starting from models with as few as 10K parameters, and evaluate downstream performance across 9 language understanding tasks. We find that scaling laws emerge at finetuning time in some NLP tasks, and that the"},"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":"2202.06387","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-02-13T19:13:00Z","cross_cats_sorted":["cs.LG","cs.NA","math.NA"],"title_canon_sha256":"4eb6be33dfdaf0345fbdcbbef45f753da9e34044a8c77914398369b29d6df5b6","abstract_canon_sha256":"f9031b705fcd125a7395e99283d247196b30b7f12c72dc2b9c4902466e20d903"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:07:38.942017Z","signature_b64":"iIJ//ZLQHka4SzI6EDlNy7zMHBdkCfxzOARCf4B2yEwpfMTt+aIcllOkstqR9ezcSGqWG6RV2o4p/02z6W9WDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f3a87150dafc1ff108567a836dae5c71d7905070724876d69de1de05d5326f6","last_reissued_at":"2026-07-05T05:07:38.941662Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:07:38.941662Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Laws Under the Microscope: Predicting Transformer Performance from Small Scale Experiments","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NA","math.NA"],"primary_cat":"cs.CL","authors_text":"Jonathan Berant, Maor Ivgi, Yair Carmon","submitted_at":"2022-02-13T19:13:00Z","abstract_excerpt":"Neural scaling laws define a predictable relationship between a model's parameter count and its performance after training in the form of a power law. However, most research to date has not explicitly investigated whether scaling laws can be used to accelerate model development. In this work, we perform such an empirical investigation across a wide range of language understanding tasks, starting from models with as few as 10K parameters, and evaluate downstream performance across 9 language understanding tasks. We find that scaling laws emerge at finetuning time in some NLP tasks, and that the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.06387","kind":"arxiv","version":2},"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/2202.06387/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":"2202.06387","created_at":"2026-07-05T05:07:38.941718+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.06387v2","created_at":"2026-07-05T05:07:38.941718+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.06387","created_at":"2026-07-05T05:07:38.941718+00:00"},{"alias_kind":"pith_short_12","alias_value":"J45IOFINV7A7","created_at":"2026-07-05T05:07:38.941718+00:00"},{"alias_kind":"pith_short_16","alias_value":"J45IOFINV7A76EEF","created_at":"2026-07-05T05:07:38.941718+00:00"},{"alias_kind":"pith_short_8","alias_value":"J45IOFIN","created_at":"2026-07-05T05:07:38.941718+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/J45IOFINV7A76EEFM6UDNWXFY4","json":"https://pith.science/pith/J45IOFINV7A76EEFM6UDNWXFY4.json","graph_json":"https://pith.science/api/pith-number/J45IOFINV7A76EEFM6UDNWXFY4/graph.json","events_json":"https://pith.science/api/pith-number/J45IOFINV7A76EEFM6UDNWXFY4/events.json","paper":"https://pith.science/paper/J45IOFIN"},"agent_actions":{"view_html":"https://pith.science/pith/J45IOFINV7A76EEFM6UDNWXFY4","download_json":"https://pith.science/pith/J45IOFINV7A76EEFM6UDNWXFY4.json","view_paper":"https://pith.science/paper/J45IOFIN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.06387&json=true","fetch_graph":"https://pith.science/api/pith-number/J45IOFINV7A76EEFM6UDNWXFY4/graph.json","fetch_events":"https://pith.science/api/pith-number/J45IOFINV7A76EEFM6UDNWXFY4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J45IOFINV7A76EEFM6UDNWXFY4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J45IOFINV7A76EEFM6UDNWXFY4/action/storage_attestation","attest_author":"https://pith.science/pith/J45IOFINV7A76EEFM6UDNWXFY4/action/author_attestation","sign_citation":"https://pith.science/pith/J45IOFINV7A76EEFM6UDNWXFY4/action/citation_signature","submit_replication":"https://pith.science/pith/J45IOFINV7A76EEFM6UDNWXFY4/action/replication_record"}},"created_at":"2026-07-05T05:07:38.941718+00:00","updated_at":"2026-07-05T05:07:38.941718+00:00"}