{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:N42RQLVK5FKXEPUG5YQJLWTMRN","short_pith_number":"pith:N42RQLVK","schema_version":"1.0","canonical_sha256":"6f35182eaae955723e86ee2095da6c8b65b25aacbf07bbe6060e84ee525207d1","source":{"kind":"arxiv","id":"2501.10573","version":1},"attestation_state":"computed","paper":{"title":"The Geometry of Tokens in Internal Representations of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Alberto Cazzaniga, Giada Panerai, Karthik Viswanathan, Matteo Biagetti, Yuri Gardinazzi","submitted_at":"2025-01-17T22:02:17Z","abstract_excerpt":"We investigate the relationship between the geometry of token embeddings and their role in the next token prediction within transformer models. An important aspect of this connection uses the notion of empirical measure, which encodes the distribution of token point clouds across transformer layers and drives the evolution of token representations in the mean-field interacting picture. We use metrics such as intrinsic dimension, neighborhood overlap, and cosine similarity to observationally probe these empirical measures across layers. To validate our approach, we compare these metrics to a da"},"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":"2501.10573","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-17T22:02:17Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"25506abbbb9a31009bd63b6271c2301dcbf606e2541447cfc9c5c65947173312","abstract_canon_sha256":"60ae251c9dccbcd914e26e94017db49af226de2448ef42ea430ac0c60f5e47a7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:30.218961Z","signature_b64":"cUw/b+BoRepVKX6RwIoOrV95ibjEoZiEk3Uu6GOcIShB+ipUMPK0TfKP2xYJM8kgGMjwLUOuIXebxVlRkxcLDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6f35182eaae955723e86ee2095da6c8b65b25aacbf07bbe6060e84ee525207d1","last_reissued_at":"2026-07-05T10:02:30.218230Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:30.218230Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Geometry of Tokens in Internal Representations of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Alberto Cazzaniga, Giada Panerai, Karthik Viswanathan, Matteo Biagetti, Yuri Gardinazzi","submitted_at":"2025-01-17T22:02:17Z","abstract_excerpt":"We investigate the relationship between the geometry of token embeddings and their role in the next token prediction within transformer models. An important aspect of this connection uses the notion of empirical measure, which encodes the distribution of token point clouds across transformer layers and drives the evolution of token representations in the mean-field interacting picture. We use metrics such as intrinsic dimension, neighborhood overlap, and cosine similarity to observationally probe these empirical measures across layers. To validate our approach, we compare these metrics to a da"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.10573","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/2501.10573/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":"2501.10573","created_at":"2026-07-05T10:02:30.218324+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.10573v1","created_at":"2026-07-05T10:02:30.218324+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.10573","created_at":"2026-07-05T10:02:30.218324+00:00"},{"alias_kind":"pith_short_12","alias_value":"N42RQLVK5FKX","created_at":"2026-07-05T10:02:30.218324+00:00"},{"alias_kind":"pith_short_16","alias_value":"N42RQLVK5FKXEPUG","created_at":"2026-07-05T10:02:30.218324+00:00"},{"alias_kind":"pith_short_8","alias_value":"N42RQLVK","created_at":"2026-07-05T10:02:30.218324+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.29462","citing_title":"MIRROR: Aligning Semantic Relations from Language to Image via Gromov--Wasserstein","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20276","citing_title":"Rethinking Intrinsic Dimension Estimation in Neural Representations","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/N42RQLVK5FKXEPUG5YQJLWTMRN","json":"https://pith.science/pith/N42RQLVK5FKXEPUG5YQJLWTMRN.json","graph_json":"https://pith.science/api/pith-number/N42RQLVK5FKXEPUG5YQJLWTMRN/graph.json","events_json":"https://pith.science/api/pith-number/N42RQLVK5FKXEPUG5YQJLWTMRN/events.json","paper":"https://pith.science/paper/N42RQLVK"},"agent_actions":{"view_html":"https://pith.science/pith/N42RQLVK5FKXEPUG5YQJLWTMRN","download_json":"https://pith.science/pith/N42RQLVK5FKXEPUG5YQJLWTMRN.json","view_paper":"https://pith.science/paper/N42RQLVK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.10573&json=true","fetch_graph":"https://pith.science/api/pith-number/N42RQLVK5FKXEPUG5YQJLWTMRN/graph.json","fetch_events":"https://pith.science/api/pith-number/N42RQLVK5FKXEPUG5YQJLWTMRN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/N42RQLVK5FKXEPUG5YQJLWTMRN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/N42RQLVK5FKXEPUG5YQJLWTMRN/action/storage_attestation","attest_author":"https://pith.science/pith/N42RQLVK5FKXEPUG5YQJLWTMRN/action/author_attestation","sign_citation":"https://pith.science/pith/N42RQLVK5FKXEPUG5YQJLWTMRN/action/citation_signature","submit_replication":"https://pith.science/pith/N42RQLVK5FKXEPUG5YQJLWTMRN/action/replication_record"}},"created_at":"2026-07-05T10:02:30.218324+00:00","updated_at":"2026-07-05T10:02:30.218324+00:00"}